code
stringlengths 87
55.2k
| code_codestyle
int64 0
349
| style_context
stringlengths 135
49.1k
| style_context_codestyle
int64 0
349
| label
int64 0
1
|
|---|---|---|---|---|
'''simple docstring'''
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A ( unittest.TestCase ):
def __init__(self : List[str] , __a : Union[str, Any] , __a : List[str]=13 , __a : List[Any]=7 , __a : Union[str, Any]=True , __a : Any=True , __a : List[Any]=True , __a : str=True , __a : List[Any]=99 , __a : Any=32 , __a : Tuple=5 , __a : Union[str, Any]=4 , __a : List[str]=37 , __a : Tuple="gelu" , __a : int=0.1 , __a : str=0.1 , __a : List[Any]=512 , __a : Optional[int]=16 , __a : Tuple=2 , __a : Any=0.02 , __a : Union[str, Any]=4 , ):
UpperCAmelCase_ = parent
UpperCAmelCase_ = batch_size
UpperCAmelCase_ = seq_length
UpperCAmelCase_ = is_training
UpperCAmelCase_ = use_attention_mask
UpperCAmelCase_ = use_token_type_ids
UpperCAmelCase_ = use_labels
UpperCAmelCase_ = vocab_size
UpperCAmelCase_ = hidden_size
UpperCAmelCase_ = num_hidden_layers
UpperCAmelCase_ = num_attention_heads
UpperCAmelCase_ = intermediate_size
UpperCAmelCase_ = hidden_act
UpperCAmelCase_ = hidden_dropout_prob
UpperCAmelCase_ = attention_probs_dropout_prob
UpperCAmelCase_ = max_position_embeddings
UpperCAmelCase_ = type_vocab_size
UpperCAmelCase_ = type_sequence_label_size
UpperCAmelCase_ = initializer_range
UpperCAmelCase_ = num_choices
def _lowercase (self : int ):
UpperCAmelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCAmelCase_ = None
if self.use_attention_mask:
UpperCAmelCase_ = random_attention_mask([self.batch_size, self.seq_length] )
UpperCAmelCase_ = None
if self.use_token_type_ids:
UpperCAmelCase_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
UpperCAmelCase_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=__a , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def _lowercase (self : List[Any] ):
UpperCAmelCase_ = self.prepare_config_and_inputs()
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ = config_and_inputs
UpperCAmelCase_ = {"input_ids": input_ids, "token_type_ids": token_type_ids, "attention_mask": attention_mask}
return config, inputs_dict
@require_flax
class __A ( UpperCamelCase__ , unittest.TestCase ):
a__ : Dict = True
a__ : Dict = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def _lowercase (self : Dict ):
UpperCAmelCase_ = FlaxRoFormerModelTester(self )
@slow
def _lowercase (self : Optional[Any] ):
for model_class_name in self.all_model_classes:
UpperCAmelCase_ = model_class_name.from_pretrained("junnyu/roformer_chinese_small" , from_pt=__a )
UpperCAmelCase_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(__a )
@require_flax
class __A ( unittest.TestCase ):
@slow
def _lowercase (self : List[Any] ):
UpperCAmelCase_ = FlaxRoFormerForMaskedLM.from_pretrained("junnyu/roformer_chinese_base" )
UpperCAmelCase_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
UpperCAmelCase_ = model(__a )[0]
UpperCAmelCase_ = 50000
UpperCAmelCase_ = (1, 6, vocab_size)
self.assertEqual(output.shape , __a )
UpperCAmelCase_ = jnp.array(
[[[-0.12_05, -1.02_65, 0.29_22], [-1.51_34, 0.19_74, 0.15_19], [-5.01_35, -3.90_03, -0.84_04]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , __a , atol=1E-4 ) )
| 1 |
"""simple docstring"""
import argparse
import logging
import os
import re
import tensorflow as tf
from transformers import (
AutoConfig,
AutoTokenizer,
DataCollatorForLanguageModeling,
PushToHubCallback,
TFAutoModelForMaskedLM,
create_optimizer,
)
__SCREAMING_SNAKE_CASE : List[str] = logging.getLogger(__name__)
__SCREAMING_SNAKE_CASE : str = tf.data.AUTOTUNE
def _a ( ) -> List[str]:
snake_case_ = argparse.ArgumentParser(description="""Train a masked language model on TPU.""" )
parser.add_argument(
"""--pretrained_model_config""" , type=_SCREAMING_SNAKE_CASE , default="""roberta-base""" , help="""The model config to use. Note that we don't copy the model's weights, only the config!""" , )
parser.add_argument(
"""--tokenizer""" , type=_SCREAMING_SNAKE_CASE , default="""unigram-tokenizer-wikitext""" , help="""The name of the tokenizer to load. We use the pretrained tokenizer to initialize the model's vocab size.""" , )
parser.add_argument(
"""--per_replica_batch_size""" , type=_SCREAMING_SNAKE_CASE , default=8 , help="""Batch size per TPU core.""" , )
parser.add_argument(
"""--no_tpu""" , action="""store_true""" , help="""If set, run on CPU and don't try to initialize a TPU. Useful for debugging on non-TPU instances.""" , )
parser.add_argument(
"""--tpu_name""" , type=_SCREAMING_SNAKE_CASE , help="""Name of TPU resource to initialize. Should be blank on Colab, and 'local' on TPU VMs.""" , default="""local""" , )
parser.add_argument(
"""--tpu_zone""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud zone that TPU resource is located in. Only used for non-Colab TPU nodes.""" , )
parser.add_argument(
"""--gcp_project""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud project name. Only used for non-Colab TPU nodes.""" )
parser.add_argument(
"""--bfloat16""" , action="""store_true""" , help="""Use mixed-precision bfloat16 for training. This is the recommended lower-precision format for TPU.""" , )
parser.add_argument(
"""--train_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to training dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--shuffle_buffer_size""" , type=_SCREAMING_SNAKE_CASE , default=2**18 , help="""Size of the shuffle buffer (in samples)""" , )
parser.add_argument(
"""--eval_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to evaluation dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--num_epochs""" , type=_SCREAMING_SNAKE_CASE , default=1 , help="""Number of epochs to train for.""" , )
parser.add_argument(
"""--learning_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-4 , help="""Learning rate to use for training.""" , )
parser.add_argument(
"""--weight_decay_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-3 , help="""Weight decay rate to use for training.""" , )
parser.add_argument(
"""--max_length""" , type=_SCREAMING_SNAKE_CASE , default=512 , help="""Maximum length of tokenized sequences. Should match the setting used in prepare_tfrecord_shards.py""" , )
parser.add_argument(
"""--mlm_probability""" , type=_SCREAMING_SNAKE_CASE , default=0.15 , help="""Fraction of tokens to mask during training.""" , )
parser.add_argument("""--output_dir""" , type=_SCREAMING_SNAKE_CASE , required=_SCREAMING_SNAKE_CASE , help="""Path to save model checkpoints to.""" )
parser.add_argument("""--hub_model_id""" , type=_SCREAMING_SNAKE_CASE , help="""Model ID to upload to on the Hugging Face Hub.""" )
snake_case_ = parser.parse_args()
return args
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
try:
if args.tpu_name:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver(
args.tpu_name , zone=args.tpu_zone , project=args.gcp_project )
else:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
raise RuntimeError(
"""Couldn't connect to TPU! Most likely you need to specify --tpu_name, --tpu_zone, or """
"""--gcp_project. When running on a TPU VM, use --tpu_name local.""" )
tf.config.experimental_connect_to_cluster(_SCREAMING_SNAKE_CASE )
tf.tpu.experimental.initialize_tpu_system(_SCREAMING_SNAKE_CASE )
return tpu
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = 0
for file in file_list:
snake_case_ = file.split("""/""" )[-1]
snake_case_ = re.search(r"""-\d+-(\d+)\.tfrecord""" , _SCREAMING_SNAKE_CASE ).group(1 )
snake_case_ = int(_SCREAMING_SNAKE_CASE )
num_samples += sample_count
return num_samples
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None ) -> Union[str, Any]:
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.data.Dataset.from_tensor_slices(_SCREAMING_SNAKE_CASE )
if shuffle:
snake_case_ = dataset.shuffle(len(_SCREAMING_SNAKE_CASE ) )
snake_case_ = tf.data.TFRecordDataset(_SCREAMING_SNAKE_CASE , num_parallel_reads=_SCREAMING_SNAKE_CASE )
# TF can't infer the total sample count because it doesn't read all the records yet, so we assert it here
snake_case_ = dataset.apply(tf.data.experimental.assert_cardinality(_SCREAMING_SNAKE_CASE ) )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
if shuffle:
assert shuffle_buffer_size is not None
snake_case_ = dataset.shuffle(args.shuffle_buffer_size )
snake_case_ = dataset.batch(_SCREAMING_SNAKE_CASE , drop_remainder=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.prefetch(_SCREAMING_SNAKE_CASE )
return dataset
def _a ( _SCREAMING_SNAKE_CASE ) -> List[Any]:
if not args.no_tpu:
snake_case_ = initialize_tpu(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.distribute.TPUStrategy(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = tf.distribute.OneDeviceStrategy(device="""/gpu:0""" )
if args.bfloataa:
tf.keras.mixed_precision.set_global_policy("""mixed_bfloat16""" )
snake_case_ = AutoTokenizer.from_pretrained(args.tokenizer )
snake_case_ = AutoConfig.from_pretrained(args.pretrained_model_config )
snake_case_ = tokenizer.vocab_size
snake_case_ = tf.io.gfile.glob(os.path.join(args.train_dataset , """*.tfrecord""" ) )
if not training_records:
raise ValueError(f"""No .tfrecord files found in {args.train_dataset}.""" )
snake_case_ = tf.io.gfile.glob(os.path.join(args.eval_dataset , """*.tfrecord""" ) )
if not eval_records:
raise ValueError(f"""No .tfrecord files found in {args.eval_dataset}.""" )
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = num_train_samples // (args.per_replica_batch_size * strategy.num_replicas_in_sync)
snake_case_ = steps_per_epoch * args.num_epochs
with strategy.scope():
snake_case_ = TFAutoModelForMaskedLM.from_config(_SCREAMING_SNAKE_CASE )
model(model.dummy_inputs ) # Pass some dummy inputs through the model to ensure all the weights are built
snake_case_ , snake_case_ = create_optimizer(
num_train_steps=_SCREAMING_SNAKE_CASE , num_warmup_steps=total_train_steps // 20 , init_lr=args.learning_rate , weight_decay_rate=args.weight_decay_rate , )
# Transformers models compute the right loss for their task by default when labels are passed, and will
# use this for training unless you specify your own loss function in compile().
model.compile(optimizer=_SCREAMING_SNAKE_CASE , metrics=["""accuracy"""] )
def decode_fn(_SCREAMING_SNAKE_CASE ):
snake_case_ = {
"""input_ids""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
"""attention_mask""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
}
return tf.io.parse_single_example(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# Many of the data collators in Transformers are TF-compilable when return_tensors == "tf", so we can
# use their methods in our data pipeline.
snake_case_ = DataCollatorForLanguageModeling(
tokenizer=_SCREAMING_SNAKE_CASE , mlm_probability=args.mlm_probability , mlm=_SCREAMING_SNAKE_CASE , return_tensors="""tf""" )
def mask_with_collator(_SCREAMING_SNAKE_CASE ):
# TF really needs an isin() function
snake_case_ = (
~tf.cast(batch["""attention_mask"""] , tf.bool )
| (batch["""input_ids"""] == tokenizer.cls_token_id)
| (batch["""input_ids"""] == tokenizer.sep_token_id)
)
snake_case_ , snake_case_ = data_collator.tf_mask_tokens(
batch["""input_ids"""] , vocab_size=len(_SCREAMING_SNAKE_CASE ) , mask_token_id=tokenizer.mask_token_id , special_tokens_mask=_SCREAMING_SNAKE_CASE , )
return batch
snake_case_ = args.per_replica_batch_size * strategy.num_replicas_in_sync
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , shuffle_buffer_size=args.shuffle_buffer_size , )
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , )
snake_case_ = []
if args.hub_model_id:
callbacks.append(
PushToHubCallback(output_dir=args.output_dir , hub_model_id=args.hub_model_id , tokenizer=_SCREAMING_SNAKE_CASE ) )
model.fit(
_SCREAMING_SNAKE_CASE , validation_data=_SCREAMING_SNAKE_CASE , epochs=args.num_epochs , callbacks=_SCREAMING_SNAKE_CASE , )
model.save_pretrained(args.output_dir )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : Union[str, Any] = parse_args()
main(args)
| 347 | 0 |
'''simple docstring'''
import math
from typing import Any, Callable, List, Optional, Tuple, Union
import numpy as np
import torch
from ...models import TaFilmDecoder
from ...schedulers import DDPMScheduler
from ...utils import is_onnx_available, logging, randn_tensor
if is_onnx_available():
from ..onnx_utils import OnnxRuntimeModel
from ..pipeline_utils import AudioPipelineOutput, DiffusionPipeline
from .continous_encoder import SpectrogramContEncoder
from .notes_encoder import SpectrogramNotesEncoder
lowerCamelCase : Optional[Any] = logging.get_logger(__name__) # pylint: disable=invalid-name
lowerCamelCase : Dict = 256
class __lowerCAmelCase (lowercase_ ):
'''simple docstring'''
lowerCAmelCase__ : List[Any] = ["""melgan"""]
def __init__(self : Tuple , UpperCamelCase : SpectrogramNotesEncoder , UpperCamelCase : SpectrogramContEncoder , UpperCamelCase : TaFilmDecoder , UpperCamelCase : DDPMScheduler , UpperCamelCase : OnnxRuntimeModel if is_onnx_available() else Any , ):
'''simple docstring'''
super().__init__()
# From MELGAN
lowercase__ = math.log(1E-5 ) # Matches MelGAN training.
lowercase__ = 4.0 # Largest value for most examples
lowercase__ = 128
self.register_modules(
notes_encoder=UpperCamelCase , continuous_encoder=UpperCamelCase , decoder=UpperCamelCase , scheduler=UpperCamelCase , melgan=UpperCamelCase , )
def UpperCamelCase__ (self : str , UpperCamelCase : Dict , UpperCamelCase : Tuple=(-1.0, 1.0) , UpperCamelCase : int=False ):
'''simple docstring'''
lowercase__ ,lowercase__ = output_range
if clip:
lowercase__ = torch.clip(UpperCamelCase , self.min_value , self.max_value )
# Scale to [0, 1].
lowercase__ = (features - self.min_value) / (self.max_value - self.min_value)
# Scale to [min_out, max_out].
return zero_one * (max_out - min_out) + min_out
def UpperCamelCase__ (self : Union[str, Any] , UpperCamelCase : List[str] , UpperCamelCase : Union[str, Any]=(-1.0, 1.0) , UpperCamelCase : Dict=False ):
'''simple docstring'''
lowercase__ ,lowercase__ = input_range
lowercase__ = torch.clip(UpperCamelCase , UpperCamelCase , UpperCamelCase ) if clip else outputs
# Scale to [0, 1].
lowercase__ = (outputs - min_out) / (max_out - min_out)
# Scale to [self.min_value, self.max_value].
return zero_one * (self.max_value - self.min_value) + self.min_value
def UpperCamelCase__ (self : Dict , UpperCamelCase : List[Any] , UpperCamelCase : List[str] , UpperCamelCase : Union[str, Any] ):
'''simple docstring'''
lowercase__ = input_tokens > 0
lowercase__ ,lowercase__ = self.notes_encoder(
encoder_input_tokens=UpperCamelCase , encoder_inputs_mask=UpperCamelCase )
lowercase__ ,lowercase__ = self.continuous_encoder(
encoder_inputs=UpperCamelCase , encoder_inputs_mask=UpperCamelCase )
return [(tokens_encoded, tokens_mask), (continuous_encoded, continuous_mask)]
def UpperCamelCase__ (self : Dict , UpperCamelCase : Union[str, Any] , UpperCamelCase : Optional[int] , UpperCamelCase : Any ):
'''simple docstring'''
lowercase__ = noise_time
if not torch.is_tensor(UpperCamelCase ):
lowercase__ = torch.tensor([timesteps] , dtype=torch.long , device=input_tokens.device )
elif torch.is_tensor(UpperCamelCase ) and len(timesteps.shape ) == 0:
lowercase__ = timesteps[None].to(input_tokens.device )
# broadcast to batch dimension in a way that's compatible with ONNX/Core ML
lowercase__ = timesteps * torch.ones(input_tokens.shape[0] , dtype=timesteps.dtype , device=timesteps.device )
lowercase__ = self.decoder(
encodings_and_masks=UpperCamelCase , decoder_input_tokens=UpperCamelCase , decoder_noise_time=UpperCamelCase )
return logits
@torch.no_grad()
def __call__(self : List[Any] , UpperCamelCase : List[List[int]] , UpperCamelCase : Optional[torch.Generator] = None , UpperCamelCase : int = 100 , UpperCamelCase : bool = True , UpperCamelCase : str = "numpy" , UpperCamelCase : Optional[Callable[[int, int, torch.FloatTensor], None]] = None , UpperCamelCase : int = 1 , ):
'''simple docstring'''
if (callback_steps is None) or (
callback_steps is not None and (not isinstance(UpperCamelCase , UpperCamelCase ) or callback_steps <= 0)
):
raise ValueError(
f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
f" {type(UpperCamelCase )}." )
lowercase__ = np.zeros([1, TARGET_FEATURE_LENGTH, self.n_dims] , dtype=np.floataa )
lowercase__ = np.zeros([1, 0, self.n_dims] , np.floataa )
lowercase__ = torch.ones((1, TARGET_FEATURE_LENGTH) , dtype=UpperCamelCase , device=self.device )
for i, encoder_input_tokens in enumerate(UpperCamelCase ):
if i == 0:
lowercase__ = torch.from_numpy(pred_mel[:1].copy() ).to(
device=self.device , dtype=self.decoder.dtype )
# The first chunk has no previous context.
lowercase__ = torch.zeros((1, TARGET_FEATURE_LENGTH) , dtype=UpperCamelCase , device=self.device )
else:
# The full song pipeline does not feed in a context feature, so the mask
# will be all 0s after the feature converter. Because we know we're
# feeding in a full context chunk from the previous prediction, set it
# to all 1s.
lowercase__ = ones
lowercase__ = self.scale_features(
UpperCamelCase , output_range=[-1.0, 1.0] , clip=UpperCamelCase )
lowercase__ = self.encode(
input_tokens=torch.IntTensor([encoder_input_tokens] ).to(device=self.device ) , continuous_inputs=UpperCamelCase , continuous_mask=UpperCamelCase , )
# Sample encoder_continuous_inputs shaped gaussian noise to begin loop
lowercase__ = randn_tensor(
shape=encoder_continuous_inputs.shape , generator=UpperCamelCase , device=self.device , dtype=self.decoder.dtype , )
# set step values
self.scheduler.set_timesteps(UpperCamelCase )
# Denoising diffusion loop
for j, t in enumerate(self.progress_bar(self.scheduler.timesteps ) ):
lowercase__ = self.decode(
encodings_and_masks=UpperCamelCase , input_tokens=UpperCamelCase , noise_time=t / self.scheduler.config.num_train_timesteps , )
# Compute previous output: x_t -> x_t-1
lowercase__ = self.scheduler.step(UpperCamelCase , UpperCamelCase , UpperCamelCase , generator=UpperCamelCase ).prev_sample
lowercase__ = self.scale_to_features(UpperCamelCase , input_range=[-1.0, 1.0] )
lowercase__ = mel[:1]
lowercase__ = mel.cpu().float().numpy()
lowercase__ = np.concatenate([full_pred_mel, pred_mel[:1]] , axis=1 )
# call the callback, if provided
if callback is not None and i % callback_steps == 0:
callback(UpperCamelCase , UpperCamelCase )
logger.info('''Generated segment''' , UpperCamelCase )
if output_type == "numpy" and not is_onnx_available():
raise ValueError(
'''Cannot return output in \'np\' format if ONNX is not available. Make sure to have ONNX installed or set \'output_type\' to \'mel\'.''' )
elif output_type == "numpy" and self.melgan is None:
raise ValueError(
'''Cannot return output in \'np\' format if melgan component is not defined. Make sure to define `self.melgan` or set \'output_type\' to \'mel\'.''' )
if output_type == "numpy":
lowercase__ = self.melgan(input_features=full_pred_mel.astype(np.floataa ) )
else:
lowercase__ = full_pred_mel
if not return_dict:
return (output,)
return AudioPipelineOutput(audios=UpperCamelCase )
| 2 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> float:
if density <= 0:
raise ValueError("""Impossible fluid density""" )
if bulk_modulus <= 0:
raise ValueError("""Impossible bulk modulus""" )
return (bulk_modulus / density) ** 0.5
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
'''simple docstring'''
from __future__ import annotations
from collections import namedtuple
from dataclasses import dataclass
@dataclass
class A :
__magic_name__ = 42
__magic_name__ = None
__magic_name__ = None
lowercase : Union[str, Any] = namedtuple('CoinsDistribResult', 'moves excess')
def lowerCAmelCase_ ( snake_case__ ):
'''simple docstring'''
if root is None:
return 0
# Validation
def count_nodes(snake_case__ ) -> int:
if node is None:
return 0
return count_nodes(node.left ) + count_nodes(node.right ) + 1
def count_coins(snake_case__ ) -> int:
if node is None:
return 0
return count_coins(node.left ) + count_coins(node.right ) + node.data
if count_nodes(snake_case__ ) != count_coins(snake_case__ ):
raise ValueError('''The nodes number should be same as the number of coins''' )
# Main calculation
def get_distrib(snake_case__ ) -> CoinsDistribResult:
if node is None:
return CoinsDistribResult(0 , 1 )
A, A : Optional[int] = get_distrib(node.left )
A, A : List[str] = get_distrib(node.right )
A : List[Any] = 1 - left_distrib_excess
A : int = 1 - right_distrib_excess
A : Union[str, Any] = (
left_distrib_moves
+ right_distrib_moves
+ abs(snake_case__ )
+ abs(snake_case__ )
)
A : Optional[int] = node.data - coins_to_left - coins_to_right
return CoinsDistribResult(snake_case__ , snake_case__ )
return get_distrib(snake_case__ )[0]
if __name__ == "__main__":
import doctest
doctest.testmod()
| 3 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE ) -> bool:
if num < 0:
return False
snake_case_ = num
snake_case_ = 0
while num > 0:
snake_case_ = rev_num * 10 + (num % 10)
num //= 10
return num_copy == rev_num
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
'''simple docstring'''
import secrets
from random import shuffle
from string import ascii_letters, ascii_lowercase, ascii_uppercase, digits, punctuation
def a_ ( lowerCamelCase : int = 8 ):
lowerCAmelCase = ascii_letters + digits + punctuation
return "".join(secrets.choice(lowerCamelCase ) for _ in range(lowerCamelCase ) )
def a_ ( lowerCamelCase : str , lowerCamelCase : int ):
# Password Generator = full boot with random_number, random_letters, and
# random_character FUNCTIONS
# Put your code here...
i -= len(lowerCamelCase )
lowerCAmelCase = i // 3
lowerCAmelCase = i % 3
# chars = chars_incl + random_letters(ascii_letters, i / 3 + remainder) +
# random_number(digits, i / 3) + random_characters(punctuation, i / 3)
lowerCAmelCase = (
chars_incl
+ random(lowerCamelCase , quotient + remainder )
+ random(lowerCamelCase , lowerCamelCase )
+ random(lowerCamelCase , lowerCamelCase )
)
lowerCAmelCase = list(lowerCamelCase )
shuffle(lowerCamelCase )
return "".join(lowerCamelCase )
# random is a generalised function for letters, characters and numbers
def a_ ( lowerCamelCase : str , lowerCamelCase : int ):
return "".join(secrets.choice(lowerCamelCase ) for _ in range(lowerCamelCase ) )
def a_ ( lowerCamelCase : str , lowerCamelCase : Tuple ):
pass # Put your code here...
def a_ ( lowerCamelCase : Optional[int] , lowerCamelCase : Tuple ):
pass # Put your code here...
def a_ ( lowerCamelCase : int , lowerCamelCase : Dict ):
pass # Put your code here...
def a_ ( lowerCamelCase : str , lowerCamelCase : int = 8 ):
if len(lowerCamelCase ) < min_length:
# Your Password must be at least 8 characters long
return False
lowerCAmelCase = any(char in ascii_uppercase for char in password )
lowerCAmelCase = any(char in ascii_lowercase for char in password )
lowerCAmelCase = any(char in digits for char in password )
lowerCAmelCase = any(char in punctuation for char in password )
return upper and lower and num and spec_char
# Passwords should contain UPPERCASE, lowerase
# numbers, and special characters
def a_ ( ):
lowerCAmelCase = int(input('Please indicate the max length of your password: ' ).strip() )
lowerCAmelCase = input(
'Please indicate the characters that must be in your password: ' ).strip()
print('Password generated:' , password_generator(lowerCamelCase ) )
print(
'Alternative Password generated:' , alternative_password_generator(lowerCamelCase , lowerCamelCase ) , )
print('[If you are thinking of using this passsword, You better save it.]' )
if __name__ == "__main__":
main()
| 4 |
"""simple docstring"""
import unittest
from transformers import SPIECE_UNDERLINE
from transformers.models.speechta import SpeechTaTokenizer
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.tokenization_utils import AddedToken
from ...test_tokenization_common import TokenizerTesterMixin
__SCREAMING_SNAKE_CASE : Tuple = get_tests_dir('fixtures/test_sentencepiece_bpe_char.model')
@require_sentencepiece
@require_tokenizers
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Tuple = SpeechTaTokenizer
__lowercase: int = False
__lowercase: List[str] = True
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
snake_case_ = SpeechTaTokenizer(UpperCAmelCase_ )
snake_case_ = AddedToken("""<mask>""" , lstrip=UpperCAmelCase_ , rstrip=UpperCAmelCase_ )
snake_case_ = mask_token
tokenizer.add_special_tokens({"""mask_token""": mask_token} )
tokenizer.add_tokens(["""<ctc_blank>"""] )
tokenizer.save_pretrained(self.tmpdirname )
def lowerCAmelCase ( self : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = """this is a test"""
snake_case_ = """this is a test"""
return input_text, output_text
def lowerCAmelCase ( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : Any=False , UpperCAmelCase_ : Tuple=20 , UpperCAmelCase_ : Dict=5 ) ->List[Any]:
"""simple docstring"""
snake_case_ , snake_case_ = self.get_input_output_texts(UpperCAmelCase_ )
snake_case_ = tokenizer.encode(UpperCAmelCase_ , add_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.decode(UpperCAmelCase_ , clean_up_tokenization_spaces=UpperCAmelCase_ )
return text, ids
def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = """<pad>"""
snake_case_ = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(UpperCAmelCase_ ) , UpperCAmelCase_ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(UpperCAmelCase_ ) , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-4] , """œ""" )
self.assertEqual(vocab_keys[-2] , """<mask>""" )
self.assertEqual(vocab_keys[-1] , """<ctc_blank>""" )
self.assertEqual(len(UpperCAmelCase_ ) , 81 )
def lowerCAmelCase ( self : Optional[int] ) ->int:
"""simple docstring"""
self.assertEqual(self.get_tokenizer().vocab_size , 79 )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = self.get_tokenizers(do_lower_case=UpperCAmelCase_ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
snake_case_ = ["""aaaaa bbbbbb""", """cccccccccdddddddd"""]
snake_case_ = tokenizer.add_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode("""aaaaa bbbbbb low cccccccccdddddddd l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
snake_case_ = {"""eos_token""": """>>>>|||<||<<|<<""", """pad_token""": """<<<<<|||>|>>>>|>"""}
snake_case_ = tokenizer.add_special_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size_a + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode(
""">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->List[str]:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = tokenizer.tokenize("""This is a test""" )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [SPIECE_UNDERLINE, """T""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """a""", SPIECE_UNDERLINE, """t""", """e""", """s""", """t"""] )
# fmt: on
self.assertListEqual(
tokenizer.convert_tokens_to_ids(UpperCAmelCase_ ) , [4, 32, 11, 10, 12, 4, 10, 12, 4, 7, 4, 6, 5, 12, 6] , )
snake_case_ = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """92000""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
snake_case_ = tokenizer.convert_tokens_to_ids(UpperCAmelCase_ )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [4, 30, 4, 20, 7, 12, 4, 25, 8, 13, 9, 4, 10, 9, 4, 3, 23, 4, 7, 9, 14, 4, 6, 11, 10, 12, 4, 10, 12, 4, 19, 7, 15, 12, 73, 26] )
# fmt: on
snake_case_ = tokenizer.convert_ids_to_tokens(UpperCAmelCase_ )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """<unk>""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = [
"""Transformers (formerly known as pytorch-transformers and pytorch-pretrained-bert) provides """
"""general-purpose architectures (BERT, GPT, RoBERTa, XLM, DistilBert, XLNet...) for Natural """
"""Language Understanding (NLU) and Natural Language Generation (NLG) with over thirty-two pretrained """
"""models in one hundred plus languages and deep interoperability between Jax, PyTorch and TensorFlow.""",
"""BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly """
"""conditioning on both left and right context in all layers.""",
"""The quick brown fox jumps over the lazy dog.""",
]
# fmt: off
snake_case_ = {
"""input_ids""": [
[4, 32, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 64, 19, 8, 13, 18, 5, 13, 15, 22, 4, 28, 9, 8, 20, 9, 4, 7, 12, 4, 24, 22, 6, 8, 13, 17, 11, 39, 6, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 7, 9, 14, 4, 24, 22, 6, 8, 13, 17, 11, 39, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 39, 25, 5, 13, 6, 63, 4, 24, 13, 8, 27, 10, 14, 5, 12, 4, 21, 5, 9, 5, 13, 7, 15, 39, 24, 16, 13, 24, 8, 12, 5, 4, 7, 13, 17, 11, 10, 6, 5, 17, 6, 16, 13, 5, 12, 4, 64, 40, 47, 54, 32, 23, 4, 53, 49, 32, 23, 4, 54, 8, 40, 47, 54, 32, 7, 23, 4, 69, 52, 43, 23, 4, 51, 10, 12, 6, 10, 15, 40, 5, 13, 6, 23, 4, 69, 52, 48, 5, 6, 26, 26, 26, 63, 4, 19, 8, 13, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 61, 9, 14, 5, 13, 12, 6, 7, 9, 14, 10, 9, 21, 4, 64, 48, 52, 61, 63, 4, 7, 9, 14, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 53, 5, 9, 5, 13, 7, 6, 10, 8, 9, 4, 64, 48, 52, 53, 63, 4, 20, 10, 6, 11, 4, 8, 27, 5, 13, 4, 6, 11, 10, 13, 6, 22, 39, 6, 20, 8, 4, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 4, 18, 8, 14, 5, 15, 12, 4, 10, 9, 4, 8, 9, 5, 4, 11, 16, 9, 14, 13, 5, 14, 4, 24, 15, 16, 12, 4, 15, 7, 9, 21, 16, 7, 21, 5, 12, 4, 7, 9, 14, 4, 14, 5, 5, 24, 4, 10, 9, 6, 5, 13, 8, 24, 5, 13, 7, 25, 10, 15, 10, 6, 22, 4, 25, 5, 6, 20, 5, 5, 9, 4, 58, 7, 37, 23, 4, 49, 22, 32, 8, 13, 17, 11, 4, 7, 9, 14, 4, 32, 5, 9, 12, 8, 13, 55, 15, 8, 20, 26, 2],
[4, 40, 47, 54, 32, 4, 10, 12, 4, 14, 5, 12, 10, 21, 9, 5, 14, 4, 6, 8, 4, 24, 13, 5, 39, 6, 13, 7, 10, 9, 4, 14, 5, 5, 24, 4, 25, 10, 14, 10, 13, 5, 17, 6, 10, 8, 9, 7, 15, 4, 13, 5, 24, 13, 5, 12, 5, 9, 6, 7, 6, 10, 8, 9, 12, 4, 19, 13, 8, 18, 4, 16, 9, 15, 7, 25, 5, 15, 5, 14, 4, 6, 5, 37, 6, 4, 25, 22, 4, 46, 8, 10, 9, 6, 15, 22, 4, 17, 8, 9, 14, 10, 6, 10, 8, 9, 10, 9, 21, 4, 8, 9, 4, 25, 8, 6, 11, 4, 15, 5, 19, 6, 4, 7, 9, 14, 4, 13, 10, 21, 11, 6, 4, 17, 8, 9, 6, 5, 37, 6, 4, 10, 9, 4, 7, 15, 15, 4, 15, 7, 22, 5, 13, 12, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[4, 32, 11, 5, 4, 45, 16, 10, 17, 28, 4, 25, 13, 8, 20, 9, 4, 19, 8, 37, 4, 46, 16, 18, 24, 12, 4, 8, 27, 5, 13, 4, 6, 11, 5, 4, 15, 7, 57, 22, 4, 14, 8, 21, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
],
"""attention_mask""": [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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],
]
}
# fmt: on
self.tokenizer_integration_test_util(
expected_encoding=UpperCAmelCase_ , model_name="""microsoft/speecht5_asr""" , revision="""c5ef64c71905caeccde0e4462ef3f9077224c524""" , sequences=UpperCAmelCase_ , )
| 347 | 0 |
import os
def UpperCAmelCase_ ( ) -> Union[str, Any]:
"""simple docstring"""
_lowercase =os.path.dirname(os.path.realpath(__snake_case ) )
_lowercase =os.path.join(__snake_case , '''triangle.txt''' )
with open(__snake_case ) as f:
_lowercase =f.readlines()
_lowercase =[]
for line in triangle:
_lowercase =[]
for number in line.strip().split(''' ''' ):
numbers_from_line.append(int(__snake_case ) )
a.append(__snake_case )
for i in range(1 , len(__snake_case ) ):
for j in range(len(a[i] ) ):
_lowercase =a[i - 1][j] if j != len(a[i - 1] ) else 0
_lowercase =a[i - 1][j - 1] if j > 0 else 0
a[i][j] += max(__snake_case , __snake_case )
return max(a[-1] )
if __name__ == "__main__":
print(solution())
| 5 |
"""simple docstring"""
import datasets
__SCREAMING_SNAKE_CASE : Tuple = '\\n@InProceedings{conneau2018xnli,\n author = "Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin",\n title = "XNLI: Evaluating Cross-lingual Sentence Representations",\n booktitle = "Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing",\n year = "2018",\n publisher = "Association for Computational Linguistics",\n location = "Brussels, Belgium",\n}\n'
__SCREAMING_SNAKE_CASE : Dict = '\\nXNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n'
__SCREAMING_SNAKE_CASE : List[str] = '\nComputes XNLI score which is just simple accuracy.\nArgs:\n predictions: Predicted labels.\n references: Ground truth labels.\nReturns:\n \'accuracy\': accuracy\nExamples:\n\n >>> predictions = [0, 1]\n >>> references = [0, 1]\n >>> xnli_metric = datasets.load_metric("xnli")\n >>> results = xnli_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {\'accuracy\': 1.0}\n'
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
return (preds == labels).mean()
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION)
class __A (datasets.Metric):
'''simple docstring'''
def lowerCAmelCase ( self : str ) ->Any:
"""simple docstring"""
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
"""predictions""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
"""references""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
} ) , codebase_urls=[] , reference_urls=[] , format="""numpy""" , )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any ) ->int:
"""simple docstring"""
return {"accuracy": simple_accuracy(UpperCAmelCase_ , UpperCAmelCase_ )}
| 347 | 0 |
import unittest
import numpy as np
from diffusers import LMSDiscreteScheduler, OnnxStableDiffusionInpaintPipeline
from diffusers.utils.testing_utils import (
is_onnx_available,
load_image,
nightly,
require_onnxruntime,
require_torch_gpu,
)
from ..test_pipelines_onnx_common import OnnxPipelineTesterMixin
if is_onnx_available():
import onnxruntime as ort
class __A( a , unittest.TestCase ):
# FIXME: add fast tests
pass
@nightly
@require_onnxruntime
@require_torch_gpu
class __A( unittest.TestCase ):
@property
def SCREAMING_SNAKE_CASE_ ( self ) -> List[str]:
'''simple docstring'''
return (
"CUDAExecutionProvider",
{
"gpu_mem_limit": "15000000000", # 15GB
"arena_extend_strategy": "kSameAsRequested",
},
)
@property
def SCREAMING_SNAKE_CASE_ ( self ) -> Union[str, Any]:
'''simple docstring'''
__a = ort.SessionOptions()
__a = False
return options
def SCREAMING_SNAKE_CASE_ ( self ) -> Optional[Any]:
'''simple docstring'''
__a = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/in_paint/overture-creations-5sI6fQgYIuo.png''' )
__a = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/in_paint/overture-creations-5sI6fQgYIuo_mask.png''' )
__a = OnnxStableDiffusionInpaintPipeline.from_pretrained(
'''runwayml/stable-diffusion-inpainting''' , revision='''onnx''' , safety_checker=_snake_case , feature_extractor=_snake_case , provider=self.gpu_provider , sess_options=self.gpu_options , )
pipe.set_progress_bar_config(disable=_snake_case )
__a = '''A red cat sitting on a park bench'''
__a = np.random.RandomState(0 )
__a = pipe(
prompt=_snake_case , image=_snake_case , mask_image=_snake_case , guidance_scale=7.5 , num_inference_steps=10 , generator=_snake_case , output_type='''np''' , )
__a = output.images
__a = images[0, 255:258, 255:258, -1]
assert images.shape == (1, 512, 512, 3)
__a = np.array([0.2514, 0.3007, 0.3517, 0.1790, 0.2382, 0.3167, 0.1944, 0.2273, 0.2464] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3
def SCREAMING_SNAKE_CASE_ ( self ) -> Tuple:
'''simple docstring'''
__a = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/in_paint/overture-creations-5sI6fQgYIuo.png''' )
__a = load_image(
'''https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'''
'''/in_paint/overture-creations-5sI6fQgYIuo_mask.png''' )
__a = LMSDiscreteScheduler.from_pretrained(
'''runwayml/stable-diffusion-inpainting''' , subfolder='''scheduler''' , revision='''onnx''' )
__a = OnnxStableDiffusionInpaintPipeline.from_pretrained(
'''runwayml/stable-diffusion-inpainting''' , revision='''onnx''' , scheduler=_snake_case , safety_checker=_snake_case , feature_extractor=_snake_case , provider=self.gpu_provider , sess_options=self.gpu_options , )
pipe.set_progress_bar_config(disable=_snake_case )
__a = '''A red cat sitting on a park bench'''
__a = np.random.RandomState(0 )
__a = pipe(
prompt=_snake_case , image=_snake_case , mask_image=_snake_case , guidance_scale=7.5 , num_inference_steps=20 , generator=_snake_case , output_type='''np''' , )
__a = output.images
__a = images[0, 255:258, 255:258, -1]
assert images.shape == (1, 512, 512, 3)
__a = np.array([0.0086, 0.0077, 0.0083, 0.0093, 0.0107, 0.0139, 0.0094, 0.0097, 0.0125] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-3 | 6 |
"""simple docstring"""
from ..utils import DummyObject, requires_backends
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Tuple ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : int ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Dict ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Union[str, Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[Any] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : str , **UpperCAmelCase_ : int ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : Union[str, Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[Any] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : List[str] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Optional[Any] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Optional[Any] ) ->int:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Dict ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Union[str, Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[int] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Optional[int] ) ->Any:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
| 347 | 0 |
import os
import sys
from contextlib import contextmanager
# Windows only
if os.name == "nt":
import ctypes
import msvcrt # noqa
class A ( ctypes.Structure ):
"""simple docstring"""
lowerCamelCase = [('size', ctypes.c_int), ('visible', ctypes.c_byte)]
def _snake_case( ) -> Optional[int]:
'''simple docstring'''
if os.name == "nt":
A__ = CursorInfo()
A__ = ctypes.windll.kernelaa.GetStdHandle(-11 )
ctypes.windll.kernelaa.GetConsoleCursorInfo(SCREAMING_SNAKE_CASE__ , ctypes.byref(SCREAMING_SNAKE_CASE__ ) )
A__ = False
ctypes.windll.kernelaa.SetConsoleCursorInfo(SCREAMING_SNAKE_CASE__ , ctypes.byref(SCREAMING_SNAKE_CASE__ ) )
elif os.name == "posix":
sys.stdout.write('\033[?25l' )
sys.stdout.flush()
def _snake_case( ) -> Optional[int]:
'''simple docstring'''
if os.name == "nt":
A__ = CursorInfo()
A__ = ctypes.windll.kernelaa.GetStdHandle(-11 )
ctypes.windll.kernelaa.GetConsoleCursorInfo(SCREAMING_SNAKE_CASE__ , ctypes.byref(SCREAMING_SNAKE_CASE__ ) )
A__ = True
ctypes.windll.kernelaa.SetConsoleCursorInfo(SCREAMING_SNAKE_CASE__ , ctypes.byref(SCREAMING_SNAKE_CASE__ ) )
elif os.name == "posix":
sys.stdout.write('\033[?25h' )
sys.stdout.flush()
@contextmanager
def _snake_case( ) -> Union[str, Any]:
'''simple docstring'''
try:
hide_cursor()
yield
finally:
show_cursor()
| 7 |
"""simple docstring"""
import warnings
from ...utils import logging
from .image_processing_mobilevit import MobileViTImageProcessor
__SCREAMING_SNAKE_CASE : int = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
def __init__( self : str , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->None:
"""simple docstring"""
warnings.warn(
"""The class MobileViTFeatureExtractor is deprecated and will be removed in version 5 of Transformers."""
""" Please use MobileViTImageProcessor instead.""" , UpperCAmelCase_ , )
super().__init__(*UpperCAmelCase_ , **UpperCAmelCase_ )
| 347 | 0 |
from ..utils import DummyObject, requires_backends
class snake_case_ ( metaclass=__A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = ["flax", "transformers"]
def __init__( self : Optional[int] , *_UpperCamelCase : int , **_UpperCamelCase : Tuple ) ->Any:
requires_backends(self , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : List[str] , *_UpperCamelCase : int , **_UpperCamelCase : str ) ->Any:
requires_backends(cls , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : Union[str, Any] , *_UpperCamelCase : Optional[Any] , **_UpperCamelCase : Any ) ->Dict:
requires_backends(cls , ['''flax''', '''transformers'''] )
class snake_case_ ( metaclass=__A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : Dict = ["flax", "transformers"]
def __init__( self : Any , *_UpperCamelCase : Optional[Any] , **_UpperCamelCase : Optional[int] ) ->Dict:
requires_backends(self , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : str , *_UpperCamelCase : List[Any] , **_UpperCamelCase : int ) ->str:
requires_backends(cls , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : str , *_UpperCamelCase : List[Any] , **_UpperCamelCase : Dict ) ->Union[str, Any]:
requires_backends(cls , ['''flax''', '''transformers'''] )
class snake_case_ ( metaclass=__A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : List[Any] = ["flax", "transformers"]
def __init__( self : Union[str, Any] , *_UpperCamelCase : Dict , **_UpperCamelCase : Tuple ) ->List[str]:
requires_backends(self , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : Optional[int] , *_UpperCamelCase : str , **_UpperCamelCase : str ) ->Union[str, Any]:
requires_backends(cls , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : Any , *_UpperCamelCase : Any , **_UpperCamelCase : List[str] ) ->int:
requires_backends(cls , ['''flax''', '''transformers'''] )
class snake_case_ ( metaclass=__A ):
'''simple docstring'''
SCREAMING_SNAKE_CASE : Optional[Any] = ["flax", "transformers"]
def __init__( self : Tuple , *_UpperCamelCase : str , **_UpperCamelCase : int ) ->Any:
requires_backends(self , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : List[Any] , *_UpperCamelCase : List[str] , **_UpperCamelCase : Union[str, Any] ) ->List[Any]:
requires_backends(cls , ['''flax''', '''transformers'''] )
@classmethod
def snake_case__( cls : Union[str, Any] , *_UpperCamelCase : str , **_UpperCamelCase : List[Any] ) ->List[Any]:
requires_backends(cls , ['''flax''', '''transformers'''] ) | 8 |
"""simple docstring"""
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Any:
snake_case_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
("""cls_token""", """vit.embeddings.cls_token"""),
("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight"""),
("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias"""),
("""pos_embed""", """vit.embeddings.position_embeddings"""),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("""norm.weight""", """layernorm.weight"""),
("""norm.bias""", """layernorm.bias"""),
("""pre_logits.fc.weight""", """pooler.dense.weight"""),
("""pre_logits.fc.bias""", """pooler.dense.bias"""),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
snake_case_ = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("""norm.weight""", """vit.layernorm.weight"""),
("""norm.bias""", """vit.layernorm.bias"""),
("""head.weight""", """classifier.weight"""),
("""head.bias""", """classifier.bias"""),
] )
return rename_keys
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
snake_case_ = """"""
else:
snake_case_ = """vit."""
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case_ = in_proj_weight[
: config.hidden_size, :
]
snake_case_ = in_proj_bias[: config.hidden_size]
snake_case_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case_ = in_proj_weight[
-config.hidden_size :, :
]
snake_case_ = in_proj_bias[-config.hidden_size :]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = ["""head.weight""", """head.bias"""]
for k in ignore_keys:
state_dict.pop(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> str:
snake_case_ = dct.pop(_SCREAMING_SNAKE_CASE )
snake_case_ = val
def _a ( ) -> Any:
snake_case_ = """http://images.cocodataset.org/val2017/000000039769.jpg"""
snake_case_ = Image.open(requests.get(_SCREAMING_SNAKE_CASE , stream=_SCREAMING_SNAKE_CASE ).raw )
return im
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = ViTConfig()
snake_case_ = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
snake_case_ = True
snake_case_ = int(vit_name[-12:-10] )
snake_case_ = int(vit_name[-9:-6] )
else:
snake_case_ = 1_000
snake_case_ = """huggingface/label-files"""
snake_case_ = """imagenet-1k-id2label.json"""
snake_case_ = json.load(open(hf_hub_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , repo_type="""dataset""" ) , """r""" ) )
snake_case_ = {int(_SCREAMING_SNAKE_CASE ): v for k, v in idalabel.items()}
snake_case_ = idalabel
snake_case_ = {v: k for k, v in idalabel.items()}
snake_case_ = int(vit_name[-6:-4] )
snake_case_ = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith("""tiny""" ):
snake_case_ = 192
snake_case_ = 768
snake_case_ = 12
snake_case_ = 3
elif vit_name[9:].startswith("""small""" ):
snake_case_ = 384
snake_case_ = 1_536
snake_case_ = 12
snake_case_ = 6
else:
pass
else:
if vit_name[4:].startswith("""small""" ):
snake_case_ = 768
snake_case_ = 2_304
snake_case_ = 8
snake_case_ = 8
elif vit_name[4:].startswith("""base""" ):
pass
elif vit_name[4:].startswith("""large""" ):
snake_case_ = 1_024
snake_case_ = 4_096
snake_case_ = 24
snake_case_ = 16
elif vit_name[4:].startswith("""huge""" ):
snake_case_ = 1_280
snake_case_ = 5_120
snake_case_ = 32
snake_case_ = 16
# load original model from timm
snake_case_ = timm.create_model(_SCREAMING_SNAKE_CASE , pretrained=_SCREAMING_SNAKE_CASE )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case_ = timm_model.state_dict()
if base_model:
remove_classification_head_(_SCREAMING_SNAKE_CASE )
snake_case_ = create_rename_keys(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
for src, dest in rename_keys:
rename_key(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
read_in_q_k_v(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# load HuggingFace model
if vit_name[-5:] == "in21k":
snake_case_ = ViTModel(_SCREAMING_SNAKE_CASE ).eval()
else:
snake_case_ = ViTForImageClassification(_SCREAMING_SNAKE_CASE ).eval()
model.load_state_dict(_SCREAMING_SNAKE_CASE )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
snake_case_ = DeiTImageProcessor(size=config.image_size )
else:
snake_case_ = ViTImageProcessor(size=config.image_size )
snake_case_ = image_processor(images=prepare_img() , return_tensors="""pt""" )
snake_case_ = encoding["""pixel_values"""]
snake_case_ = model(_SCREAMING_SNAKE_CASE )
if base_model:
snake_case_ = timm_model.forward_features(_SCREAMING_SNAKE_CASE )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.pooler_output , atol=1E-3 )
else:
snake_case_ = timm_model(_SCREAMING_SNAKE_CASE )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.logits , atol=1E-3 )
Path(_SCREAMING_SNAKE_CASE ).mkdir(exist_ok=_SCREAMING_SNAKE_CASE )
print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_SCREAMING_SNAKE_CASE )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--vit_name',
default='vit_base_patch16_224',
type=str,
help='Name of the ViT timm model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
__SCREAMING_SNAKE_CASE : int = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 347 | 0 |
import argparse
import gdown
import numpy as np
import torch
from huggingface_hub import hf_hub_download
from transformers import (
CLIPTokenizer,
CLIPTokenizerFast,
VideoMAEImageProcessor,
XCLIPConfig,
XCLIPModel,
XCLIPProcessor,
XCLIPTextConfig,
XCLIPVisionConfig,
)
def _UpperCamelCase ( lowercase__ , lowercase__ ):
__SCREAMING_SNAKE_CASE : List[Any] = XCLIPTextConfig()
# derive patch size from model name
__SCREAMING_SNAKE_CASE : Optional[int] = model_name.find('''patch''' )
__SCREAMING_SNAKE_CASE : Dict = int(model_name[start_idx + len('''patch''' ) : start_idx + len('''patch''' ) + 2] )
__SCREAMING_SNAKE_CASE : Union[str, Any] = XCLIPVisionConfig(patch_size=lowercase__ , num_frames=lowercase__ )
if "large" in model_name:
__SCREAMING_SNAKE_CASE : int = 768
__SCREAMING_SNAKE_CASE : Dict = 3072
__SCREAMING_SNAKE_CASE : List[Any] = 12
__SCREAMING_SNAKE_CASE : Any = 1024
__SCREAMING_SNAKE_CASE : Union[str, Any] = 4096
__SCREAMING_SNAKE_CASE : Union[str, Any] = 16
__SCREAMING_SNAKE_CASE : List[Any] = 24
__SCREAMING_SNAKE_CASE : Union[str, Any] = 768
__SCREAMING_SNAKE_CASE : str = 3072
if model_name == "xclip-large-patch14-16-frames":
__SCREAMING_SNAKE_CASE : Any = 336
__SCREAMING_SNAKE_CASE : int = XCLIPConfig.from_text_vision_configs(lowercase__ , lowercase__ )
if "large" in model_name:
__SCREAMING_SNAKE_CASE : Optional[int] = 768
return config
def _UpperCamelCase ( lowercase__ ):
# text encoder
if name == "token_embedding.weight":
__SCREAMING_SNAKE_CASE : Dict = name.replace('''token_embedding.weight''' , '''text_model.embeddings.token_embedding.weight''' )
if name == "positional_embedding":
__SCREAMING_SNAKE_CASE : Optional[Any] = name.replace('''positional_embedding''' , '''text_model.embeddings.position_embedding.weight''' )
if "ln_1" in name:
__SCREAMING_SNAKE_CASE : Dict = name.replace('''ln_1''' , '''layer_norm1''' )
if "ln_2" in name:
__SCREAMING_SNAKE_CASE : Optional[int] = name.replace('''ln_2''' , '''layer_norm2''' )
if "c_fc" in name:
__SCREAMING_SNAKE_CASE : Tuple = name.replace('''c_fc''' , '''fc1''' )
if "c_proj" in name:
__SCREAMING_SNAKE_CASE : Tuple = name.replace('''c_proj''' , '''fc2''' )
if name.startswith('''transformer.resblocks''' ):
__SCREAMING_SNAKE_CASE : int = name.replace('''transformer.resblocks''' , '''text_model.encoder.layers''' )
if "attn.out_proj" in name and "message" not in name:
__SCREAMING_SNAKE_CASE : List[Any] = name.replace('''attn.out_proj''' , '''self_attn.out_proj''' )
if "ln_final" in name:
__SCREAMING_SNAKE_CASE : List[Any] = name.replace('''ln_final''' , '''text_model.final_layer_norm''' )
# visual encoder
if name == "visual.class_embedding":
__SCREAMING_SNAKE_CASE : List[str] = name.replace('''visual.class_embedding''' , '''vision_model.embeddings.class_embedding''' )
if name == "visual.positional_embedding":
__SCREAMING_SNAKE_CASE : str = name.replace('''visual.positional_embedding''' , '''vision_model.embeddings.position_embedding.weight''' )
if name.startswith('''visual.transformer.resblocks''' ):
__SCREAMING_SNAKE_CASE : List[str] = name.replace('''visual.transformer.resblocks''' , '''vision_model.encoder.layers''' )
if "visual.conv1" in name:
__SCREAMING_SNAKE_CASE : List[Any] = name.replace('''visual.conv1''' , '''vision_model.embeddings.patch_embedding''' )
if "visual.ln_pre" in name:
__SCREAMING_SNAKE_CASE : Dict = name.replace('''visual.ln_pre''' , '''vision_model.pre_layernorm''' )
if "visual.ln_post" in name:
__SCREAMING_SNAKE_CASE : Optional[Any] = name.replace('''visual.ln_post''' , '''vision_model.post_layernorm''' )
if "visual.proj" in name:
__SCREAMING_SNAKE_CASE : str = name.replace('''visual.proj''' , '''visual_projection.weight''' )
if "text_projection" in name:
__SCREAMING_SNAKE_CASE : List[str] = name.replace('''text_projection''' , '''text_projection.weight''' )
# things on top
if "prompts_visual_proj" in name:
__SCREAMING_SNAKE_CASE : Tuple = name.replace('''prompts_visual_proj''' , '''prompts_visual_projection''' )
if "prompts_visual_ln" in name:
__SCREAMING_SNAKE_CASE : Any = name.replace('''prompts_visual_ln''' , '''prompts_visual_layernorm''' )
# mit
if name == "mit.positional_embedding":
__SCREAMING_SNAKE_CASE : Dict = name.replace('''positional''' , '''position''' )
if name.startswith('''mit.resblocks''' ):
__SCREAMING_SNAKE_CASE : Optional[int] = name.replace('''mit.resblocks''' , '''mit.encoder.layers''' )
# prompts generator
if name.startswith('''prompts_generator.norm''' ):
__SCREAMING_SNAKE_CASE : List[Any] = name.replace('''prompts_generator.norm''' , '''prompts_generator.layernorm''' )
return name
def _UpperCamelCase ( lowercase__ , lowercase__ ):
for key in orig_state_dict.copy().keys():
__SCREAMING_SNAKE_CASE : List[str] = orig_state_dict.pop(lowercase__ )
if "attn.in_proj" in key:
__SCREAMING_SNAKE_CASE : str = key.split('''.''' )
if key.startswith('''visual''' ):
__SCREAMING_SNAKE_CASE : str = key_split[3]
__SCREAMING_SNAKE_CASE : Union[str, Any] = config.vision_config.hidden_size
if "message_attn" in key:
if "weight" in key:
__SCREAMING_SNAKE_CASE : int = val[
:dim, :
]
__SCREAMING_SNAKE_CASE : str = val[
dim : dim * 2, :
]
__SCREAMING_SNAKE_CASE : str = val[
-dim:, :
]
else:
__SCREAMING_SNAKE_CASE : str = val[
:dim
]
__SCREAMING_SNAKE_CASE : Union[str, Any] = val[
dim : dim * 2
]
__SCREAMING_SNAKE_CASE : Optional[int] = val[
-dim:
]
else:
if "weight" in key:
__SCREAMING_SNAKE_CASE : Any = val[
:dim, :
]
__SCREAMING_SNAKE_CASE : Optional[int] = val[
dim : dim * 2, :
]
__SCREAMING_SNAKE_CASE : Optional[int] = val[
-dim:, :
]
else:
__SCREAMING_SNAKE_CASE : Optional[int] = val[:dim]
__SCREAMING_SNAKE_CASE : Optional[int] = val[
dim : dim * 2
]
__SCREAMING_SNAKE_CASE : Tuple = val[-dim:]
elif key.startswith('''mit''' ):
__SCREAMING_SNAKE_CASE : Dict = key_split[2]
__SCREAMING_SNAKE_CASE : Any = config.vision_config.mit_hidden_size
if "weight" in key:
__SCREAMING_SNAKE_CASE : Dict = val[:dim, :]
__SCREAMING_SNAKE_CASE : Union[str, Any] = val[dim : dim * 2, :]
__SCREAMING_SNAKE_CASE : List[Any] = val[-dim:, :]
else:
__SCREAMING_SNAKE_CASE : Dict = val[:dim]
__SCREAMING_SNAKE_CASE : Optional[Any] = val[dim : dim * 2]
__SCREAMING_SNAKE_CASE : List[str] = val[-dim:]
else:
__SCREAMING_SNAKE_CASE : Optional[Any] = key_split[2]
__SCREAMING_SNAKE_CASE : Optional[int] = config.text_config.hidden_size
if "weight" in key:
__SCREAMING_SNAKE_CASE : Dict = val[:dim, :]
__SCREAMING_SNAKE_CASE : Optional[Any] = val[
dim : dim * 2, :
]
__SCREAMING_SNAKE_CASE : Optional[Any] = val[-dim:, :]
else:
__SCREAMING_SNAKE_CASE : Tuple = val[:dim]
__SCREAMING_SNAKE_CASE : Any = val[
dim : dim * 2
]
__SCREAMING_SNAKE_CASE : Any = val[-dim:]
else:
__SCREAMING_SNAKE_CASE : List[str] = rename_key(lowercase__ )
if new_key_name in ["visual_projection.weight", "text_projection.weight"]:
__SCREAMING_SNAKE_CASE : Any = val.T
__SCREAMING_SNAKE_CASE : Dict = val
return orig_state_dict
def _UpperCamelCase ( lowercase__ ):
if num_frames == 8:
__SCREAMING_SNAKE_CASE : List[str] = '''eating_spaghetti_8_frames.npy'''
elif num_frames == 16:
__SCREAMING_SNAKE_CASE : str = '''eating_spaghetti.npy'''
elif num_frames == 32:
__SCREAMING_SNAKE_CASE : Dict = '''eating_spaghetti_32_frames.npy'''
__SCREAMING_SNAKE_CASE : str = hf_hub_download(
repo_id='''hf-internal-testing/spaghetti-video''' , filename=lowercase__ , repo_type='''dataset''' , )
__SCREAMING_SNAKE_CASE : Union[str, Any] = np.load(lowercase__ )
return list(lowercase__ )
def _UpperCamelCase ( lowercase__ , lowercase__=None , lowercase__=False ):
__SCREAMING_SNAKE_CASE : Union[str, Any] = {
# fully supervised kinetics-400 checkpoints
'''xclip-base-patch32''': '''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_8.pth''',
'''xclip-base-patch32-16-frames''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_16.pth'''
),
'''xclip-base-patch16''': '''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_8.pth''',
'''xclip-base-patch16-16-frames''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_16.pth'''
),
'''xclip-large-patch14''': '''https://drive.google.com/u/0/uc?id=1NUOImq0o5DlQTST17iIP3vG7DgmHQuCx&export=download&confirm=t&uuid=b26caedc-88e2-473e-830a-9d158b653cdb''',
'''xclip-large-patch14-16-frames''': '''https://drive.google.com/u/0/uc?id=1FOYgnJc097OJ4lGwtRCCydQyVPJEOH7d&export=download&confirm=t&uuid=538fa810-e671-4050-b385-9a623f89804f''',
# fully supervised kinetics-600 checkpoints
'''xclip-base-patch16-kinetics-600''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_8.pth'''
),
'''xclip-base-patch16-kinetics-600-16-frames''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_16.pth'''
),
'''xclip-large-patch14-kinetics-600''': '''https://drive.google.com/u/0/uc?id=1FV8C1INuM91sLAN4ImjzePLIlpMSihwV&export=download&confirm=t&uuid=141d4977-4a65-44ae-864f-4b0c19f838be''',
# few shot
'''xclip-base-patch16-hmdb-2-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_2.pth'''
),
'''xclip-base-patch16-hmdb-4-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_4.pth'''
),
'''xclip-base-patch16-hmdb-8-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_8.pth'''
),
'''xclip-base-patch16-hmdb-16-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_16.pth'''
),
'''xclip-base-patch16-ucf-2-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_2.pth'''
),
'''xclip-base-patch16-ucf-4-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_4.pth'''
),
'''xclip-base-patch16-ucf-8-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_8.pth'''
),
'''xclip-base-patch16-ucf-16-shot''': (
'''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_16.pth'''
),
# zero shot
'''xclip-base-patch16-zero-shot''': '''https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/zero.pth''',
}
__SCREAMING_SNAKE_CASE : Optional[Any] = model_to_url[model_name]
__SCREAMING_SNAKE_CASE : int = 8
if "16-frames" in model_name:
__SCREAMING_SNAKE_CASE : Tuple = 16
elif "shot" in model_name:
__SCREAMING_SNAKE_CASE : List[Any] = 32
__SCREAMING_SNAKE_CASE : Any = get_xclip_config(lowercase__ , lowercase__ )
__SCREAMING_SNAKE_CASE : str = XCLIPModel(lowercase__ )
model.eval()
if "drive" in checkpoint_url:
__SCREAMING_SNAKE_CASE : Optional[int] = '''pytorch_model.bin'''
gdown.cached_download(lowercase__ , lowercase__ , quiet=lowercase__ )
__SCREAMING_SNAKE_CASE : Optional[int] = torch.load(lowercase__ , map_location='''cpu''' )['''model''']
else:
__SCREAMING_SNAKE_CASE : List[str] = torch.hub.load_state_dict_from_url(lowercase__ )['''model''']
__SCREAMING_SNAKE_CASE : str = convert_state_dict(lowercase__ , lowercase__ )
__SCREAMING_SNAKE_CASE : Dict = XCLIPModel(lowercase__ )
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE : Any = model.load_state_dict(lowercase__ , strict=lowercase__ )
assert missing_keys == ["text_model.embeddings.position_ids", "vision_model.embeddings.position_ids"]
model.eval()
__SCREAMING_SNAKE_CASE : Tuple = 336 if model_name == '''xclip-large-patch14-16-frames''' else 224
__SCREAMING_SNAKE_CASE : Dict = VideoMAEImageProcessor(size=lowercase__ )
__SCREAMING_SNAKE_CASE : List[Any] = CLIPTokenizer.from_pretrained('''openai/clip-vit-base-patch32''' )
__SCREAMING_SNAKE_CASE : int = CLIPTokenizerFast.from_pretrained('''openai/clip-vit-base-patch32''' )
__SCREAMING_SNAKE_CASE : Optional[Any] = XCLIPProcessor(image_processor=lowercase__ , tokenizer=lowercase__ )
__SCREAMING_SNAKE_CASE : Dict = prepare_video(lowercase__ )
__SCREAMING_SNAKE_CASE : int = processor(
text=['''playing sports''', '''eating spaghetti''', '''go shopping'''] , videos=lowercase__ , return_tensors='''pt''' , padding=lowercase__ )
print('''Shape of pixel values:''' , inputs.pixel_values.shape )
with torch.no_grad():
__SCREAMING_SNAKE_CASE : Optional[Any] = model(**lowercase__ )
# Verify outputs
__SCREAMING_SNAKE_CASE : Tuple = outputs.logits_per_video
__SCREAMING_SNAKE_CASE : Optional[int] = logits_per_video.softmax(dim=1 )
print('''Probs:''' , lowercase__ )
# kinetics-400
if model_name == "xclip-base-patch32":
__SCREAMING_SNAKE_CASE : Union[str, Any] = torch.tensor([[0.0019, 0.9951, 0.0030]] )
elif model_name == "xclip-base-patch32-16-frames":
__SCREAMING_SNAKE_CASE : Dict = torch.tensor([[7.09_99e-04, 9.98_83e-01, 4.55_80e-04]] )
elif model_name == "xclip-base-patch16":
__SCREAMING_SNAKE_CASE : Optional[Any] = torch.tensor([[0.0083, 0.9681, 0.0236]] )
elif model_name == "xclip-base-patch16-16-frames":
__SCREAMING_SNAKE_CASE : Optional[int] = torch.tensor([[7.69_37e-04, 9.97_28e-01, 1.94_73e-03]] )
elif model_name == "xclip-large-patch14":
__SCREAMING_SNAKE_CASE : Any = torch.tensor([[0.0062, 0.9864, 0.0075]] )
elif model_name == "xclip-large-patch14-16-frames":
__SCREAMING_SNAKE_CASE : Union[str, Any] = torch.tensor([[3.38_77e-04, 9.99_37e-01, 2.88_88e-04]] )
# kinetics-600
elif model_name == "xclip-base-patch16-kinetics-600":
__SCREAMING_SNAKE_CASE : Any = torch.tensor([[0.0555, 0.8914, 0.0531]] )
elif model_name == "xclip-base-patch16-kinetics-600-16-frames":
__SCREAMING_SNAKE_CASE : Union[str, Any] = torch.tensor([[3.85_54e-04, 9.99_29e-01, 3.27_54e-04]] )
elif model_name == "xclip-large-patch14-kinetics-600":
__SCREAMING_SNAKE_CASE : List[Any] = torch.tensor([[0.0036, 0.9920, 0.0045]] )
# few shot
elif model_name == "xclip-base-patch16-hmdb-2-shot":
__SCREAMING_SNAKE_CASE : str = torch.tensor([[7.18_90e-06, 9.99_94e-01, 5.65_59e-05]] )
elif model_name == "xclip-base-patch16-hmdb-4-shot":
__SCREAMING_SNAKE_CASE : Dict = torch.tensor([[1.03_20e-05, 9.99_93e-01, 6.24_35e-05]] )
elif model_name == "xclip-base-patch16-hmdb-8-shot":
__SCREAMING_SNAKE_CASE : Union[str, Any] = torch.tensor([[4.13_77e-06, 9.99_90e-01, 9.83_86e-05]] )
elif model_name == "xclip-base-patch16-hmdb-16-shot":
__SCREAMING_SNAKE_CASE : Any = torch.tensor([[4.13_47e-05, 9.99_62e-01, 3.34_11e-04]] )
elif model_name == "xclip-base-patch16-ucf-2-shot":
__SCREAMING_SNAKE_CASE : Dict = torch.tensor([[8.58_57e-05, 9.99_28e-01, 6.32_91e-04]] )
elif model_name == "xclip-base-patch16-ucf-4-shot":
__SCREAMING_SNAKE_CASE : Optional[Any] = torch.tensor([[8.58_57e-05, 9.99_28e-01, 6.32_91e-04]] )
elif model_name == "xclip-base-patch16-ucf-8-shot":
__SCREAMING_SNAKE_CASE : List[Any] = torch.tensor([[0.0027, 0.9904, 0.0070]] )
elif model_name == "xclip-base-patch16-ucf-16-shot":
__SCREAMING_SNAKE_CASE : str = torch.tensor([[9.82_19e-04, 9.95_93e-01, 3.08_63e-03]] )
# zero shot
elif model_name == "xclip-base-patch16-zero-shot":
__SCREAMING_SNAKE_CASE : List[Any] = torch.tensor([[3.50_82e-04, 9.97_85e-01, 1.79_66e-03]] )
else:
raise ValueError(F'''Model name {model_name} not supported''' )
assert torch.allclose(lowercase__ , lowercase__ , atol=1e-3 )
print('''Looks ok!''' )
if pytorch_dump_folder_path is not None:
print(F'''Saving model {model_name} to {pytorch_dump_folder_path}''' )
model.save_pretrained(lowercase__ )
if push_to_hub:
print('''Pushing model, processor and slow tokenizer files to the hub...''' )
model.push_to_hub(lowercase__ , organization='''nielsr''' )
processor.push_to_hub(lowercase__ , organization='''nielsr''' )
slow_tokenizer.push_to_hub(lowercase__ , organization='''nielsr''' )
if __name__ == "__main__":
__lowerCAmelCase : Union[str, Any] =argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--model_name',
default='xclip-base-patch32',
type=str,
help='Name of the model.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
parser.add_argument(
'--push_to_hub', action='store_true', help='Whether or not to push the converted model to the 🤗 hub.'
)
__lowerCAmelCase : Optional[int] =parser.parse_args()
convert_xclip_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
| 9 |
"""simple docstring"""
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A (unittest.TestCase):
'''simple docstring'''
def __init__( self : List[Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Tuple=13 , UpperCAmelCase_ : List[Any]=7 , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=99 , UpperCAmelCase_ : str=32 , UpperCAmelCase_ : Tuple=5 , UpperCAmelCase_ : Union[str, Any]=4 , UpperCAmelCase_ : Any=37 , UpperCAmelCase_ : int="gelu" , UpperCAmelCase_ : Any=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Dict=512 , UpperCAmelCase_ : Optional[Any]=16 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : str=0.02 , UpperCAmelCase_ : str=4 , ) ->Tuple:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_attention_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = type_vocab_size
snake_case_ = type_sequence_label_size
snake_case_ = initializer_range
snake_case_ = num_choices
def lowerCAmelCase ( self : Optional[int] ) ->str:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_attention_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = None
if self.use_token_type_ids:
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=UpperCAmelCase_ , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def lowerCAmelCase ( self : List[str] ) ->Dict:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """token_type_ids""": token_type_ids, """attention_mask""": attention_mask}
return config, inputs_dict
@require_flax
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Union[str, Any] = True
__lowercase: int = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = FlaxRoFormerModelTester(self )
@slow
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""junnyu/roformer_chinese_small""" , from_pt=UpperCAmelCase_ )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
@require_flax
class __A (unittest.TestCase):
'''simple docstring'''
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = FlaxRoFormerForMaskedLM.from_pretrained("""junnyu/roformer_chinese_base""" )
snake_case_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
snake_case_ = model(UpperCAmelCase_ )[0]
snake_case_ = 50_000
snake_case_ = (1, 6, vocab_size)
self.assertEqual(output.shape , UpperCAmelCase_ )
snake_case_ = jnp.array(
[[[-0.1_205, -1.0_265, 0.2_922], [-1.5_134, 0.1_974, 0.1_519], [-5.0_135, -3.9_003, -0.8_404]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=1E-4 ) )
| 347 | 0 |
import collections
import os
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
from ...tokenization_utils import PreTrainedTokenizer
from ...utils import logging
__A = logging.get_logger(__name__)
__A = "▁"
__A = {"vocab_file": "prophetnet.tokenizer"}
__A = {
"vocab_file": {
"microsoft/xprophetnet-large-wiki100-cased": (
"https://huggingface.co/microsoft/xprophetnet-large-wiki100-cased/resolve/main/prophetnet.tokenizer"
),
}
}
__A = {
"microsoft/xprophetnet-large-wiki100-cased": {"do_lower_case": False},
}
__A = {
"microsoft/xprophetnet-large-wiki100-cased": 512,
}
def lowerCAmelCase_ ( __a ) -> int:
"""simple docstring"""
lowerCamelCase__: Optional[Any] =collections.OrderedDict()
with open(__a , "r" , encoding="utf-8" ) as reader:
lowerCamelCase__: int =reader.readlines()
for index, token in enumerate(__a ):
lowerCamelCase__: List[str] =token.rstrip("\n" )
lowerCamelCase__: List[Any] =index
return vocab
class _SCREAMING_SNAKE_CASE ( __SCREAMING_SNAKE_CASE ):
'''simple docstring'''
lowercase_ = VOCAB_FILES_NAMES
lowercase_ = PRETRAINED_VOCAB_FILES_MAP
lowercase_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
lowercase_ = ["input_ids", "attention_mask"]
def __init__(self : Union[str, Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[Any]="[SEP]" , UpperCAmelCase_ : List[Any]="[SEP]" , UpperCAmelCase_ : Optional[Any]="[SEP]" , UpperCAmelCase_ : int="[UNK]" , UpperCAmelCase_ : Optional[Any]="[PAD]" , UpperCAmelCase_ : Dict="[CLS]" , UpperCAmelCase_ : Dict="[MASK]" , UpperCAmelCase_ : Optional[Dict[str, Any]] = None , **UpperCAmelCase_ : Tuple , ) ->None:
'''simple docstring'''
lowerCamelCase__: int ={} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
bos_token=UpperCAmelCase_ , eos_token=UpperCAmelCase_ , sep_token=UpperCAmelCase_ , unk_token=UpperCAmelCase_ , pad_token=UpperCAmelCase_ , cls_token=UpperCAmelCase_ , mask_token=UpperCAmelCase_ , sp_model_kwargs=self.sp_model_kwargs , **UpperCAmelCase_ , )
try:
import sentencepiece as spm
except ImportError:
logger.warning(
"You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece"
" pip install sentencepiece")
raise
lowerCamelCase__: Optional[int] =spm.SentencePieceProcessor(**self.sp_model_kwargs)
self.sp_model.Load(str(UpperCAmelCase_))
lowerCamelCase__: Optional[int] =vocab_file
# Original fairseq vocab and spm vocab must be "aligned":
# Vocab | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9
# -------- | ------- | ------- | ------ | ------- | --- | --- | --- | ----- | ----- | ----
# fairseq | '<s>' | '<pad>' | '</s>' | '<unk>' | ',' | '.' | '▁' | 's' | '▁de' | '-'
# spm | '<unk>' | '<s>' | '</s>' | ',' | '.' | '▁' | 's' | '▁de' | '-' | '▁a'
# put special tokens and [unused] tokens into the vocab
lowerCamelCase__: Optional[int] ={"[PAD]": 0, "[CLS]": 1, "[SEP]": 2, "[UNK]": 3, "[MASK]": 4}
for i in range(10):
lowerCamelCase__: Optional[int] =F"""[unused{i}]"""
lowerCamelCase__: int =5 + i
# The first "real" token "," has position 15 in the embedding vocab and position 3 in the spm vocab
lowerCamelCase__: int =12
lowerCamelCase__: Optional[Any] ={v: k for k, v in self.fairseq_tokens_to_ids.items()}
for k in self.fairseq_tokens_to_ids.keys():
self.unique_no_split_tokens.append(UpperCAmelCase_)
def __getstate__(self : List[str]) ->Dict:
'''simple docstring'''
lowerCamelCase__: Optional[int] =self.__dict__.copy()
lowerCamelCase__: Dict =None
return state
def __setstate__(self : List[str] , UpperCAmelCase_ : Union[str, Any]) ->Dict:
'''simple docstring'''
lowerCamelCase__: Tuple =d
try:
import sentencepiece as spm
except ImportError:
logger.warning(
"You need to install SentencePiece to use XLMRobertaTokenizer: https://github.com/google/sentencepiece"
" pip install sentencepiece")
raise
# for backward compatibility
if not hasattr(self , "sp_model_kwargs"):
lowerCamelCase__: Dict ={}
lowerCamelCase__: Tuple =spm.SentencePieceProcessor(**self.sp_model_kwargs)
self.sp_model.Load(self.vocab_file)
def SCREAMING_SNAKE_CASE_ (self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None , UpperCAmelCase_ : bool = False) ->List[int]:
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCAmelCase_ , token_ids_a=UpperCAmelCase_ , already_has_special_tokens=UpperCAmelCase_)
if token_ids_a is None:
return ([0] * len(UpperCAmelCase_)) + [1]
return ([0] * len(UpperCAmelCase_)) + [1] + ([0] * len(UpperCAmelCase_)) + [1]
def SCREAMING_SNAKE_CASE_ (self : Dict , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None) ->List[int]:
'''simple docstring'''
lowerCamelCase__: Any =[self.sep_token_id]
if token_ids_a is None:
return len(token_ids_a + sep) * [0]
return len(token_ids_a + sep + sep + token_ids_a + sep) * [0]
@property
def SCREAMING_SNAKE_CASE_ (self : str) ->Dict:
'''simple docstring'''
return len(self.sp_model) + self.fairseq_offset
def SCREAMING_SNAKE_CASE_ (self : List[str]) ->Tuple:
'''simple docstring'''
lowerCamelCase__: str ={self.convert_ids_to_tokens(UpperCAmelCase_): i for i in range(self.vocab_size)}
vocab.update(self.added_tokens_encoder)
return vocab
def SCREAMING_SNAKE_CASE_ (self : Union[str, Any] , UpperCAmelCase_ : str) ->str:
'''simple docstring'''
return self.sp_model.encode(UpperCAmelCase_ , out_type=UpperCAmelCase_)
def SCREAMING_SNAKE_CASE_ (self : List[Any] , UpperCAmelCase_ : List[Any]) ->str:
'''simple docstring'''
if token in self.fairseq_tokens_to_ids:
return self.fairseq_tokens_to_ids[token]
lowerCamelCase__: str =self.sp_model.PieceToId(UpperCAmelCase_)
# Need to return unknown token if the SP model returned 0
return spm_id + self.fairseq_offset if spm_id else self.unk_token_id
def SCREAMING_SNAKE_CASE_ (self : str , UpperCAmelCase_ : Optional[Any]) ->Optional[int]:
'''simple docstring'''
if index in self.fairseq_ids_to_tokens:
return self.fairseq_ids_to_tokens[index]
return self.sp_model.IdToPiece(index - self.fairseq_offset)
def SCREAMING_SNAKE_CASE_ (self : int , UpperCAmelCase_ : Optional[Any]) ->Optional[Any]:
'''simple docstring'''
lowerCamelCase__: Union[str, Any] ="".join(UpperCAmelCase_).replace(UpperCAmelCase_ , " ").strip()
return out_string
def SCREAMING_SNAKE_CASE_ (self : Union[str, Any] , UpperCAmelCase_ : str , UpperCAmelCase_ : Optional[str] = None) ->Tuple[str]:
'''simple docstring'''
if not os.path.isdir(UpperCAmelCase_):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""")
return
lowerCamelCase__: List[str] =os.path.join(
UpperCAmelCase_ , (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"])
if os.path.abspath(self.vocab_file) != os.path.abspath(UpperCAmelCase_) and os.path.isfile(self.vocab_file):
copyfile(self.vocab_file , UpperCAmelCase_)
elif not os.path.isfile(self.vocab_file):
with open(UpperCAmelCase_ , "wb") as fi:
lowerCamelCase__: Dict =self.sp_model.serialized_model_proto()
fi.write(UpperCAmelCase_)
return (out_vocab_file,)
def SCREAMING_SNAKE_CASE_ (self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None) ->List[int]:
'''simple docstring'''
if token_ids_a is None:
return token_ids_a + [self.sep_token_id]
lowerCamelCase__: Union[str, Any] =[self.sep_token_id]
return token_ids_a + sep + token_ids_a + sep
| 10 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> bool:
snake_case_ = get_failure_array(_SCREAMING_SNAKE_CASE )
# 2) Step through text searching for pattern
snake_case_ , snake_case_ = 0, 0 # index into text, pattern
while i < len(_SCREAMING_SNAKE_CASE ):
if pattern[j] == text[i]:
if j == (len(_SCREAMING_SNAKE_CASE ) - 1):
return True
j += 1
# if this is a prefix in our pattern
# just go back far enough to continue
elif j > 0:
snake_case_ = failure[j - 1]
continue
i += 1
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> list[int]:
snake_case_ = [0]
snake_case_ = 0
snake_case_ = 1
while j < len(_SCREAMING_SNAKE_CASE ):
if pattern[i] == pattern[j]:
i += 1
elif i > 0:
snake_case_ = failure[i - 1]
continue
j += 1
failure.append(_SCREAMING_SNAKE_CASE )
return failure
if __name__ == "__main__":
# Test 1)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abc1abc12'
__SCREAMING_SNAKE_CASE : Optional[int] = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
__SCREAMING_SNAKE_CASE : List[str] = 'alskfjaldsk23adsfabcabc'
assert kmp(pattern, texta) and not kmp(pattern, texta)
# Test 2)
__SCREAMING_SNAKE_CASE : int = 'ABABX'
__SCREAMING_SNAKE_CASE : Optional[Any] = 'ABABZABABYABABX'
assert kmp(pattern, text)
# Test 3)
__SCREAMING_SNAKE_CASE : Any = 'AAAB'
__SCREAMING_SNAKE_CASE : List[Any] = 'ABAAAAAB'
assert kmp(pattern, text)
# Test 4)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abcdabcy'
__SCREAMING_SNAKE_CASE : str = 'abcxabcdabxabcdabcdabcy'
assert kmp(pattern, text)
# Test 5)
__SCREAMING_SNAKE_CASE : Any = 'aabaabaaa'
assert get_failure_array(pattern) == [0, 1, 0, 1, 2, 3, 4, 5, 2]
| 347 | 0 |
import warnings
from typing import List, Optional, Union
from ...processing_utils import ProcessorMixin
from ...tokenization_utils_base import BatchEncoding, PaddingStrategy, PreTokenizedInput, TextInput, TruncationStrategy
from ...utils import TensorType
class lowerCAmelCase__ ( a):
'''simple docstring'''
__SCREAMING_SNAKE_CASE = ["image_processor", "tokenizer"]
__SCREAMING_SNAKE_CASE = "LayoutLMv3ImageProcessor"
__SCREAMING_SNAKE_CASE = ("LayoutLMv3Tokenizer", "LayoutLMv3TokenizerFast")
def __init__( self , __lowerCamelCase=None , __lowerCamelCase=None , **__lowerCamelCase) -> Any:
_A : Optional[Any] = None
if "feature_extractor" in kwargs:
warnings.warn(
"The `feature_extractor` argument is deprecated and will be removed in v5, use `image_processor`"
" instead." , __lowerCamelCase , )
_A : List[str] = kwargs.pop("feature_extractor")
_A : Union[str, Any] = image_processor if image_processor is not None else feature_extractor
if image_processor is None:
raise ValueError("You need to specify an `image_processor`.")
if tokenizer is None:
raise ValueError("You need to specify a `tokenizer`.")
super().__init__(__lowerCamelCase , __lowerCamelCase)
def __call__( self , __lowerCamelCase , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = True , __lowerCamelCase = False , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = 0 , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = None , __lowerCamelCase = False , __lowerCamelCase = False , __lowerCamelCase = False , __lowerCamelCase = False , __lowerCamelCase = True , __lowerCamelCase = None , **__lowerCamelCase , ) -> BatchEncoding:
# verify input
if self.image_processor.apply_ocr and (boxes is not None):
raise ValueError(
"You cannot provide bounding boxes if you initialized the image processor with apply_ocr set to True.")
if self.image_processor.apply_ocr and (word_labels is not None):
raise ValueError(
"You cannot provide word labels if you initialized the image processor with apply_ocr set to True.")
# first, apply the image processor
_A : List[Any] = self.image_processor(images=__lowerCamelCase , return_tensors=__lowerCamelCase)
# second, apply the tokenizer
if text is not None and self.image_processor.apply_ocr and text_pair is None:
if isinstance(__lowerCamelCase , __lowerCamelCase):
_A : Dict = [text] # add batch dimension (as the image processor always adds a batch dimension)
_A : Tuple = features["words"]
_A : List[str] = self.tokenizer(
text=text if text is not None else features["words"] , text_pair=text_pair if text_pair is not None else None , boxes=boxes if boxes is not None else features["boxes"] , word_labels=__lowerCamelCase , add_special_tokens=__lowerCamelCase , padding=__lowerCamelCase , truncation=__lowerCamelCase , max_length=__lowerCamelCase , stride=__lowerCamelCase , pad_to_multiple_of=__lowerCamelCase , return_token_type_ids=__lowerCamelCase , return_attention_mask=__lowerCamelCase , return_overflowing_tokens=__lowerCamelCase , return_special_tokens_mask=__lowerCamelCase , return_offsets_mapping=__lowerCamelCase , return_length=__lowerCamelCase , verbose=__lowerCamelCase , return_tensors=__lowerCamelCase , **__lowerCamelCase , )
# add pixel values
_A : Dict = features.pop("pixel_values")
if return_overflowing_tokens is True:
_A : int = self.get_overflowing_images(__lowerCamelCase , encoded_inputs["overflow_to_sample_mapping"])
_A : int = images
return encoded_inputs
def _lowerCamelCase ( self , __lowerCamelCase , __lowerCamelCase) -> Tuple:
# in case there's an overflow, ensure each `input_ids` sample is mapped to its corresponding image
_A : Optional[Any] = []
for sample_idx in overflow_to_sample_mapping:
images_with_overflow.append(images[sample_idx])
if len(__lowerCamelCase) != len(__lowerCamelCase):
raise ValueError(
"Expected length of images to be the same as the length of `overflow_to_sample_mapping`, but got"
F" {len(__lowerCamelCase)} and {len(__lowerCamelCase)}")
return images_with_overflow
def _lowerCamelCase ( self , *__lowerCamelCase , **__lowerCamelCase) -> List[str]:
return self.tokenizer.batch_decode(*__lowerCamelCase , **__lowerCamelCase)
def _lowerCamelCase ( self , *__lowerCamelCase , **__lowerCamelCase) -> List[str]:
return self.tokenizer.decode(*__lowerCamelCase , **__lowerCamelCase)
@property
def _lowerCamelCase ( self) -> Optional[Any]:
return ["input_ids", "bbox", "attention_mask", "pixel_values"]
@property
def _lowerCamelCase ( self) -> str:
warnings.warn(
"`feature_extractor_class` is deprecated and will be removed in v5. Use `image_processor_class` instead." , __lowerCamelCase , )
return self.image_processor_class
@property
def _lowerCamelCase ( self) -> List[Any]:
warnings.warn(
"`feature_extractor` is deprecated and will be removed in v5. Use `image_processor` instead." , __lowerCamelCase , )
return self.image_processor
| 11 |
"""simple docstring"""
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __A (snake_case__):
'''simple docstring'''
@slow
@require_torch
def lowerCAmelCase ( self : Union[str, Any] ) ->Dict:
"""simple docstring"""
snake_case_ = EncoderDecoderModel.from_encoder_decoder_pretrained("""prajjwal1/bert-tiny""" , """prajjwal1/bert-tiny""" )
snake_case_ = BertTokenizer.from_pretrained("""bert-base-uncased""" )
snake_case_ = bertabert.config.encoder.vocab_size
snake_case_ = tokenizer.sep_token_id
snake_case_ = tokenizer.cls_token_id
snake_case_ = 128
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""train[:1%]""" )
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""validation[:1%]""" )
snake_case_ = train_dataset.select(range(32 ) )
snake_case_ = val_dataset.select(range(16 ) )
snake_case_ = 4
def _map_to_encoder_decoder_inputs(UpperCAmelCase_ : int ):
# Tokenizer will automatically set [BOS] <text> [EOS]
snake_case_ = tokenizer(batch["""article"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=512 )
snake_case_ = tokenizer(batch["""highlights"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=128 )
snake_case_ = inputs.input_ids
snake_case_ = inputs.attention_mask
snake_case_ = outputs.input_ids
snake_case_ = outputs.input_ids.copy()
snake_case_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch["""labels"""]
]
snake_case_ = outputs.attention_mask
assert all(len(UpperCAmelCase_ ) == 512 for x in inputs.input_ids )
assert all(len(UpperCAmelCase_ ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(UpperCAmelCase_ : Union[str, Any] ):
snake_case_ = pred.label_ids
snake_case_ = pred.predictions
# all unnecessary tokens are removed
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(UpperCAmelCase_ ) )] ) / len(UpperCAmelCase_ )
return {"accuracy": accuracy}
# map train dataset
snake_case_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
train_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
# same for validation dataset
snake_case_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
val_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
snake_case_ = self.get_auto_remove_tmp_dir()
snake_case_ = SeqaSeqTrainingArguments(
output_dir=UpperCAmelCase_ , per_device_train_batch_size=UpperCAmelCase_ , per_device_eval_batch_size=UpperCAmelCase_ , predict_with_generate=UpperCAmelCase_ , evaluation_strategy="""steps""" , do_train=UpperCAmelCase_ , do_eval=UpperCAmelCase_ , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
snake_case_ = SeqaSeqTrainer(
model=UpperCAmelCase_ , args=UpperCAmelCase_ , compute_metrics=_compute_metrics , train_dataset=UpperCAmelCase_ , eval_dataset=UpperCAmelCase_ , tokenizer=UpperCAmelCase_ , )
# start training
trainer.train()
| 347 | 0 |
from collections import OrderedDict
from typing import Any, Mapping, Optional
from ... import PreTrainedTokenizer, TensorType, is_torch_available
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfigWithPast
from ...utils import logging
UpperCAmelCase_ = logging.get_logger(__name__)
UpperCAmelCase_ = {
'EleutherAI/gpt-neo-1.3B': 'https://huggingface.co/EleutherAI/gpt-neo-1.3B/resolve/main/config.json',
# See all GPTNeo models at https://huggingface.co/models?filter=gpt_neo
}
class lowerCamelCase__( __lowerCamelCase):
UpperCAmelCase__ : int = 'gpt_neo'
UpperCAmelCase__ : Union[str, Any] = ['past_key_values']
UpperCAmelCase__ : int = {'num_attention_heads': 'num_heads', 'num_hidden_layers': 'num_layers'}
def __init__( self: List[Any] , UpperCamelCase_: Union[str, Any]=5_02_57 , UpperCamelCase_: Any=20_48 , UpperCamelCase_: Optional[Any]=20_48 , UpperCamelCase_: Any=24 , UpperCamelCase_: int=[[["global", "local"], 12]] , UpperCamelCase_: Optional[int]=16 , UpperCamelCase_: str=None , UpperCamelCase_: Any=2_56 , UpperCamelCase_: List[str]="gelu_new" , UpperCamelCase_: Union[str, Any]=0.0 , UpperCamelCase_: Optional[Any]=0.0 , UpperCamelCase_: Union[str, Any]=0.0 , UpperCamelCase_: Dict=0.1 , UpperCamelCase_: List[Any]=1E-5 , UpperCamelCase_: Optional[Any]=0.02 , UpperCamelCase_: int=True , UpperCamelCase_: Optional[Any]=5_02_56 , UpperCamelCase_: Dict=5_02_56 , **UpperCamelCase_: int , ):
__lowerCamelCase = vocab_size
__lowerCamelCase = max_position_embeddings
__lowerCamelCase = hidden_size
__lowerCamelCase = num_layers
__lowerCamelCase = num_heads
__lowerCamelCase = intermediate_size
__lowerCamelCase = window_size
__lowerCamelCase = activation_function
__lowerCamelCase = resid_dropout
__lowerCamelCase = embed_dropout
__lowerCamelCase = attention_dropout
__lowerCamelCase = classifier_dropout
__lowerCamelCase = layer_norm_epsilon
__lowerCamelCase = initializer_range
__lowerCamelCase = use_cache
__lowerCamelCase = bos_token_id
__lowerCamelCase = eos_token_id
__lowerCamelCase = attention_types
__lowerCamelCase = self.expand_attention_types_params(UpperCamelCase_ )
if len(self.attention_layers ) != self.num_layers:
raise ValueError(
"""Configuration for convolutional module is incorrect. """
"""It is required that `len(config.attention_layers)` == `config.num_layers` """
F'but is `len(config.attention_layers) = {len(self.attention_layers )}`, '
F'`config.num_layers = {self.num_layers}`. '
"""`config.attention_layers` is prepared using `config.attention_types`. """
"""Please verify the value of `config.attention_types` argument.""" )
super().__init__(bos_token_id=UpperCamelCase_ , eos_token_id=UpperCamelCase_ , **UpperCamelCase_ )
@staticmethod
def lowerCAmelCase__ ( UpperCamelCase_: Tuple ):
__lowerCamelCase = []
for item in attention_types:
for _ in range(item[1] ):
attentions.extend(item[0] )
return attentions
def lowerCamelCase__ ( A__ : int , A__ : Optional[Any] , A__ : Optional[int] , A__ : List[Any] ):
'''simple docstring'''
import torch
__lowerCamelCase = input.size()
__lowerCamelCase = len(A__ )
__lowerCamelCase = shape[dimension]
__lowerCamelCase = torch.arange(0 , A__ , A__ )
__lowerCamelCase = torch.div(sizedim - size , A__ , rounding_mode="""floor""" ) + 1
__lowerCamelCase = torch.arange(A__ ) + low_indices[:min_length][:, None]
__lowerCamelCase = [slice(A__ )] * rank
__lowerCamelCase = indices
__lowerCamelCase = input[s]
__lowerCamelCase = list(range(0 , rank + 1 ) )
perm.append(perm.pop(dimension + 1 ) )
return sliced.permute(A__ )
def lowerCamelCase__ ( A__ : Optional[int] , A__ : List[str] ):
'''simple docstring'''
import torch
__lowerCamelCase = torch.arange(1 , A__ )
__lowerCamelCase = torch.remainder(A__ , A__ )
__lowerCamelCase = remainders == 0
__lowerCamelCase = candidates[divisor_indices]
__lowerCamelCase = torch.max(A__ )
return largest_divisor, torch.div(A__ , A__ , rounding_mode="""floor""" )
class lowerCamelCase__( __lowerCamelCase):
@property
def lowerCAmelCase__ ( self: Any ):
__lowerCamelCase = OrderedDict({"""input_ids""": {0: """batch""", 1: """sequence"""}} )
if self.use_past:
self.fill_with_past_key_values_(UpperCamelCase_ , direction="""inputs""" )
__lowerCamelCase = {0: """batch""", 1: """past_sequence + sequence"""}
else:
__lowerCamelCase = {0: """batch""", 1: """sequence"""}
return common_inputs
@property
def lowerCAmelCase__ ( self: Dict ):
return self._config.num_heads
def lowerCAmelCase__ ( self: Tuple , UpperCamelCase_: PreTrainedTokenizer , UpperCamelCase_: int = -1 , UpperCamelCase_: int = -1 , UpperCamelCase_: bool = False , UpperCamelCase_: Optional[TensorType] = None , ):
__lowerCamelCase = super(UpperCamelCase_ , self ).generate_dummy_inputs(
UpperCamelCase_ , batch_size=UpperCamelCase_ , seq_length=UpperCamelCase_ , is_pair=UpperCamelCase_ , framework=UpperCamelCase_ )
# We need to order the input in the way they appears in the forward()
__lowerCamelCase = OrderedDict({"""input_ids""": common_inputs["""input_ids"""]} )
# Need to add the past_keys
if self.use_past:
if not is_torch_available():
raise ValueError("""Cannot generate dummy past_keys inputs without PyTorch installed.""" )
else:
import torch
__lowerCamelCase, __lowerCamelCase = common_inputs["""input_ids"""].shape
# Not using the same length for past_key_values
__lowerCamelCase = seqlen + 2
__lowerCamelCase = (
batch,
self.num_attention_heads,
past_key_values_length,
self._config.hidden_size // self.num_attention_heads,
)
__lowerCamelCase = [
(torch.zeros(UpperCamelCase_ ), torch.zeros(UpperCamelCase_ )) for _ in range(self.num_layers )
]
__lowerCamelCase = common_inputs["""attention_mask"""]
if self.use_past:
__lowerCamelCase = ordered_inputs["""attention_mask"""].dtype
__lowerCamelCase = torch.cat(
[ordered_inputs["""attention_mask"""], torch.ones(UpperCamelCase_ , UpperCamelCase_ , dtype=UpperCamelCase_ )] , dim=1 )
return ordered_inputs
@property
def lowerCAmelCase__ ( self: Optional[Any] ):
return 13
| 12 |
"""simple docstring"""
# this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.:
# python ./utils/get_modified_files.py utils src tests examples
#
# it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered
# since the output of this script is fed into Makefile commands it doesn't print a newline after the results
import re
import subprocess
import sys
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8')
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split()
__SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:])
__SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""")
__SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)]
print(' '.join(relevant_modified_files), end='')
| 347 | 0 |
import unittest
import numpy as np
from transformers import MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING, TF_MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING
from transformers.pipelines import AudioClassificationPipeline, pipeline
from transformers.testing_utils import (
is_pipeline_test,
nested_simplify,
require_tf,
require_torch,
require_torchaudio,
slow,
)
from .test_pipelines_common import ANY
@is_pipeline_test
class __lowercase ( unittest.TestCase ):
"""simple docstring"""
_UpperCAmelCase : Optional[int] = MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING
_UpperCAmelCase : int = TF_MODEL_FOR_AUDIO_CLASSIFICATION_MAPPING
def _SCREAMING_SNAKE_CASE ( self : List[Any] , lowerCAmelCase__ : List[str] , lowerCAmelCase__ : Tuple , lowerCAmelCase__ : Dict):
SCREAMING_SNAKE_CASE_: Any = AudioClassificationPipeline(model=lowerCAmelCase__ , feature_extractor=lowerCAmelCase__)
# test with a raw waveform
SCREAMING_SNAKE_CASE_: Union[str, Any] = np.zeros((3_4000,))
SCREAMING_SNAKE_CASE_: List[Any] = np.zeros((1_4000,))
return audio_classifier, [audioa, audio]
def _SCREAMING_SNAKE_CASE ( self : Optional[int] , lowerCAmelCase__ : List[str] , lowerCAmelCase__ : Optional[Any]):
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_: Any = examples
SCREAMING_SNAKE_CASE_: List[str] = audio_classifier(lowerCAmelCase__)
# by default a model is initialized with num_labels=2
self.assertEqual(
lowerCAmelCase__ , [
{"score": ANY(lowerCAmelCase__), "label": ANY(lowerCAmelCase__)},
{"score": ANY(lowerCAmelCase__), "label": ANY(lowerCAmelCase__)},
] , )
SCREAMING_SNAKE_CASE_: Dict = audio_classifier(lowerCAmelCase__ , top_k=1)
self.assertEqual(
lowerCAmelCase__ , [
{"score": ANY(lowerCAmelCase__), "label": ANY(lowerCAmelCase__)},
] , )
self.run_torchaudio(lowerCAmelCase__)
@require_torchaudio
def _SCREAMING_SNAKE_CASE ( self : Union[str, Any] , lowerCAmelCase__ : Dict):
import datasets
# test with a local file
SCREAMING_SNAKE_CASE_: Optional[Any] = datasets.load_dataset("hf-internal-testing/librispeech_asr_dummy" , "clean" , split="validation")
SCREAMING_SNAKE_CASE_: List[Any] = dataset[0]["audio"]["array"]
SCREAMING_SNAKE_CASE_: Tuple = audio_classifier(lowerCAmelCase__)
self.assertEqual(
lowerCAmelCase__ , [
{"score": ANY(lowerCAmelCase__), "label": ANY(lowerCAmelCase__)},
{"score": ANY(lowerCAmelCase__), "label": ANY(lowerCAmelCase__)},
] , )
@require_torch
def _SCREAMING_SNAKE_CASE ( self : int):
SCREAMING_SNAKE_CASE_: List[str] = "anton-l/wav2vec2-random-tiny-classifier"
SCREAMING_SNAKE_CASE_: Tuple = pipeline("audio-classification" , model=lowerCAmelCase__)
SCREAMING_SNAKE_CASE_: str = np.ones((8000,))
SCREAMING_SNAKE_CASE_: Optional[int] = audio_classifier(lowerCAmelCase__ , top_k=4)
SCREAMING_SNAKE_CASE_: List[Any] = [
{"score": 0.0842, "label": "no"},
{"score": 0.0838, "label": "up"},
{"score": 0.0837, "label": "go"},
{"score": 0.0834, "label": "right"},
]
SCREAMING_SNAKE_CASE_: List[Any] = [
{"score": 0.0845, "label": "stop"},
{"score": 0.0844, "label": "on"},
{"score": 0.0841, "label": "right"},
{"score": 0.0834, "label": "left"},
]
self.assertIn(nested_simplify(lowerCAmelCase__ , decimals=4) , [EXPECTED_OUTPUT, EXPECTED_OUTPUT_PT_2])
SCREAMING_SNAKE_CASE_: int = {"array": np.ones((8000,)), "sampling_rate": audio_classifier.feature_extractor.sampling_rate}
SCREAMING_SNAKE_CASE_: str = audio_classifier(lowerCAmelCase__ , top_k=4)
self.assertIn(nested_simplify(lowerCAmelCase__ , decimals=4) , [EXPECTED_OUTPUT, EXPECTED_OUTPUT_PT_2])
@require_torch
@slow
def _SCREAMING_SNAKE_CASE ( self : Any):
import datasets
SCREAMING_SNAKE_CASE_: Optional[Any] = "superb/wav2vec2-base-superb-ks"
SCREAMING_SNAKE_CASE_: Union[str, Any] = pipeline("audio-classification" , model=lowerCAmelCase__)
SCREAMING_SNAKE_CASE_: Any = datasets.load_dataset("anton-l/superb_dummy" , "ks" , split="test")
SCREAMING_SNAKE_CASE_: List[Any] = np.array(dataset[3]["speech"] , dtype=np.floataa)
SCREAMING_SNAKE_CASE_: str = audio_classifier(lowerCAmelCase__ , top_k=4)
self.assertEqual(
nested_simplify(lowerCAmelCase__ , decimals=3) , [
{"score": 0.981, "label": "go"},
{"score": 0.007, "label": "up"},
{"score": 0.006, "label": "_unknown_"},
{"score": 0.001, "label": "down"},
] , )
@require_tf
@unittest.skip("Audio classification is not implemented for TF")
def _SCREAMING_SNAKE_CASE ( self : List[str]):
pass
| 13 |
"""simple docstring"""
import argparse
import json
import os
import fairseq
import torch
from fairseq.data import Dictionary
from transformers import (
WavaVecaConfig,
WavaVecaCTCTokenizer,
WavaVecaFeatureExtractor,
WavaVecaForCTC,
WavaVecaForPreTraining,
WavaVecaProcessor,
logging,
)
from transformers.models.wavaveca.modeling_wavaveca import WavaVecaForSequenceClassification
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'post_extract_proj': 'feature_projection.projection',
'encoder.pos_conv.0': 'encoder.pos_conv_embed.conv',
'self_attn.k_proj': 'encoder.layers.*.attention.k_proj',
'self_attn.v_proj': 'encoder.layers.*.attention.v_proj',
'self_attn.q_proj': 'encoder.layers.*.attention.q_proj',
'self_attn.out_proj': 'encoder.layers.*.attention.out_proj',
'self_attn_layer_norm': 'encoder.layers.*.layer_norm',
'fc1': 'encoder.layers.*.feed_forward.intermediate_dense',
'fc2': 'encoder.layers.*.feed_forward.output_dense',
'final_layer_norm': 'encoder.layers.*.final_layer_norm',
'encoder.layer_norm': 'encoder.layer_norm',
'adapter_layer': 'encoder.layers.*.adapter_layer',
'w2v_model.layer_norm': 'feature_projection.layer_norm',
'quantizer.weight_proj': 'quantizer.weight_proj',
'quantizer.vars': 'quantizer.codevectors',
'project_q': 'project_q',
'final_proj': 'project_hid',
'w2v_encoder.proj': 'lm_head',
'mask_emb': 'masked_spec_embed',
'pooling_layer.linear': 'projector',
'pooling_layer.projection': 'classifier',
}
__SCREAMING_SNAKE_CASE : List[Any] = [
'lm_head',
'quantizer.weight_proj',
'quantizer.codevectors',
'project_q',
'project_hid',
'projector',
'classifier',
]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = {}
with open(_SCREAMING_SNAKE_CASE , """r""" ) as file:
for line_number, line in enumerate(_SCREAMING_SNAKE_CASE ):
snake_case_ = line.strip()
if line:
snake_case_ = line.split()
snake_case_ = line_number
snake_case_ = words[0]
snake_case_ = value
return result
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
for attribute in key.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ).shape
elif weight_type is not None and weight_type == "param":
snake_case_ = hf_pointer
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = shape_pointer.shape
# let's reduce dimension
snake_case_ = value[0]
else:
snake_case_ = hf_pointer.shape
if hf_shape != value.shape:
raise ValueError(
f"""Shape of hf {key + "." + weight_type if weight_type is not None else ""} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}""" )
if weight_type == "weight":
snake_case_ = value
elif weight_type == "weight_g":
snake_case_ = value
elif weight_type == "weight_v":
snake_case_ = value
elif weight_type == "bias":
snake_case_ = value
elif weight_type == "param":
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = value
else:
snake_case_ = value
logger.info(f"""{key + "." + weight_type if weight_type is not None else ""} was initialized from {full_name}.""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = """.""".join([key, weight_type] )
elif weight_type is not None and weight_type == "param":
snake_case_ = """.""".join([key, hf_param_name] )
else:
snake_case_ = key
snake_case_ = value if """lm_head""" in full_key else value[0]
__SCREAMING_SNAKE_CASE : int = {
'W_a': 'linear_1.weight',
'W_b': 'linear_2.weight',
'b_a': 'linear_1.bias',
'b_b': 'linear_2.bias',
'ln_W': 'norm.weight',
'ln_b': 'norm.bias',
}
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> List[str]:
snake_case_ = False
for key, mapped_key in MAPPING.items():
snake_case_ = """wav2vec2.""" + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key
if key in name or key.split("""w2v_model.""" )[-1] == name.split(""".""" )[0]:
snake_case_ = True
if "*" in mapped_key:
snake_case_ = name.split(_SCREAMING_SNAKE_CASE )[0].split(""".""" )[-2]
snake_case_ = mapped_key.replace("""*""" , _SCREAMING_SNAKE_CASE )
if "weight_g" in name:
snake_case_ = """weight_g"""
elif "weight_v" in name:
snake_case_ = """weight_v"""
elif "bias" in name:
snake_case_ = """bias"""
elif "weight" in name:
# TODO: don't match quantizer.weight_proj
snake_case_ = """weight"""
else:
snake_case_ = None
if hf_dict is not None:
rename_dict(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
else:
set_recursively(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
return is_used
return is_used
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = []
snake_case_ = fairseq_model.state_dict()
snake_case_ = hf_model.wavaveca.feature_extractor
for name, value in fairseq_dict.items():
snake_case_ = False
if "conv_layers" in name:
load_conv_layer(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , hf_model.config.feat_extract_norm == """group""" , )
snake_case_ = True
else:
snake_case_ = load_wavaveca_layer(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not is_used:
unused_weights.append(_SCREAMING_SNAKE_CASE )
logger.warning(f"""Unused weights: {unused_weights}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
snake_case_ = full_name.split("""conv_layers.""" )[-1]
snake_case_ = name.split(""".""" )
snake_case_ = int(items[0] )
snake_case_ = int(items[1] )
if type_id == 0:
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(_SCREAMING_SNAKE_CASE )
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=False ) -> int:
if config_path is not None:
snake_case_ = WavaVecaConfig.from_pretrained(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaConfig()
if is_seq_class:
snake_case_ = read_txt_into_dict(_SCREAMING_SNAKE_CASE )
snake_case_ = idalabel
snake_case_ = WavaVecaForSequenceClassification(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
feature_extractor.save_pretrained(_SCREAMING_SNAKE_CASE )
elif is_finetuned:
if dict_path:
snake_case_ = Dictionary.load(_SCREAMING_SNAKE_CASE )
# important change bos & pad token id since CTC symbol is <pad> and
# not <s> as in fairseq
snake_case_ = target_dict.pad_index
snake_case_ = target_dict.bos_index
snake_case_ = target_dict.eos_index
snake_case_ = len(target_dict.symbols )
snake_case_ = os.path.join(_SCREAMING_SNAKE_CASE , """vocab.json""" )
if not os.path.isdir(_SCREAMING_SNAKE_CASE ):
logger.error("""--pytorch_dump_folder_path ({}) should be a directory""".format(_SCREAMING_SNAKE_CASE ) )
return
os.makedirs(_SCREAMING_SNAKE_CASE , exist_ok=_SCREAMING_SNAKE_CASE )
snake_case_ = target_dict.indices
# fairseq has the <pad> and <s> switched
snake_case_ = 0
snake_case_ = 1
with open(_SCREAMING_SNAKE_CASE , """w""" , encoding="""utf-8""" ) as vocab_handle:
json.dump(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaCTCTokenizer(
_SCREAMING_SNAKE_CASE , unk_token=target_dict.unk_word , pad_token=target_dict.pad_word , bos_token=target_dict.bos_word , eos_token=target_dict.eos_word , word_delimiter_token="""|""" , do_lower_case=_SCREAMING_SNAKE_CASE , )
snake_case_ = True if config.feat_extract_norm == """layer""" else False
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
snake_case_ = WavaVecaProcessor(feature_extractor=_SCREAMING_SNAKE_CASE , tokenizer=_SCREAMING_SNAKE_CASE )
processor.save_pretrained(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaForCTC(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaForPreTraining(_SCREAMING_SNAKE_CASE )
if is_finetuned or is_seq_class:
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={"""data""": """/""".join(dict_path.split("""/""" )[:-1] )} )
else:
snake_case_ = argparse.Namespace(task="""audio_pretraining""" )
snake_case_ = fairseq.tasks.setup_task(_SCREAMING_SNAKE_CASE )
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] , task=_SCREAMING_SNAKE_CASE )
snake_case_ = model[0].eval()
recursively_load_weights(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , not is_finetuned )
hf_wavavec.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : str = argparse.ArgumentParser()
parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.')
parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to fairseq checkpoint')
parser.add_argument('--dict_path', default=None, type=str, help='Path to dict of fine-tuned model')
parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert')
parser.add_argument(
'--not_finetuned', action='store_true', help='Whether the model to convert is a fine-tuned model or not'
)
parser.add_argument(
'--is_seq_class',
action='store_true',
help='Whether the model to convert is a fine-tuned sequence classification model or not',
)
__SCREAMING_SNAKE_CASE : Any = parser.parse_args()
__SCREAMING_SNAKE_CASE : List[Any] = not args.not_finetuned and not args.is_seq_class
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.config_path,
args.dict_path,
is_finetuned,
args.is_seq_class,
)
| 347 | 0 |
from __future__ import annotations
def SCREAMING_SNAKE_CASE ( lowercase_ , lowercase_ , lowercase_ ) -> tuple[float, list[float]]:
"""simple docstring"""
A__ = list(range(len(lowercase_ ) ) )
A__ = [v / w for v, w in zip(lowercase_ , lowercase_ )]
index.sort(key=lambda lowercase_ : ratio[i] , reverse=lowercase_ )
A__ = 0
A__ = [0] * len(lowercase_ )
for i in index:
if weight[i] <= capacity:
A__ = 1
max_value += value[i]
capacity -= weight[i]
else:
A__ = capacity / weight[i]
max_value += value[i] * capacity / weight[i]
break
return max_value, fractions
if __name__ == "__main__":
import doctest
doctest.testmod()
| 14 |
"""simple docstring"""
import tempfile
import unittest
import numpy as np
import transformers
from transformers import GPTaTokenizer, GPTJConfig, is_flax_available, is_torch_available
from transformers.testing_utils import is_pt_flax_cross_test, require_flax, tooslow
from ...generation.test_flax_utils import FlaxGenerationTesterMixin
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.modeling_flax_pytorch_utils import (
convert_pytorch_state_dict_to_flax,
load_flax_weights_in_pytorch_model,
)
from transformers.models.gptj.modeling_flax_gptj import FlaxGPTJForCausalLM, FlaxGPTJModel
if is_torch_available():
import torch
class __A :
'''simple docstring'''
def __init__( self : Dict , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Any=14 , UpperCAmelCase_ : Union[str, Any]=7 , UpperCAmelCase_ : Tuple=True , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Union[str, Any]=False , UpperCAmelCase_ : Union[str, Any]=True , UpperCAmelCase_ : str=99 , UpperCAmelCase_ : Union[str, Any]=32 , UpperCAmelCase_ : List[Any]=4 , UpperCAmelCase_ : Optional[int]=4 , UpperCAmelCase_ : int=4 , UpperCAmelCase_ : str=37 , UpperCAmelCase_ : Any="gelu" , UpperCAmelCase_ : str=0.1 , UpperCAmelCase_ : Union[str, Any]=0.1 , UpperCAmelCase_ : int=512 , UpperCAmelCase_ : Tuple=0.02 , ) ->List[str]:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_input_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = rotary_dim
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = initializer_range
snake_case_ = None
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
def lowerCAmelCase ( self : int ) ->Optional[int]:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_input_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = GPTJConfig(
vocab_size=self.vocab_size , n_embd=self.hidden_size , n_layer=self.num_hidden_layers , n_head=self.num_attention_heads , n_positions=self.max_position_embeddings , use_cache=UpperCAmelCase_ , bos_token_id=self.bos_token_id , eos_token_id=self.eos_token_id , pad_token_id=self.pad_token_id , rotary_dim=self.rotary_dim , )
return (config, input_ids, input_mask)
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """attention_mask""": attention_mask}
return config, inputs_dict
def lowerCAmelCase ( self : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.ones((input_ids.shape[0], max_decoder_length) , dtype="""i4""" )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , attention_mask=UpperCAmelCase_ , past_key_values=outputs_cache.past_key_values , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
def lowerCAmelCase ( self : Union[str, Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = jnp.concatenate(
[attention_mask, jnp.zeros((attention_mask.shape[0], max_decoder_length - attention_mask.shape[1]) )] , axis=-1 , )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , past_key_values=outputs_cache.past_key_values , attention_mask=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
@require_flax
class __A (snake_case__ , snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Any = (FlaxGPTJModel, FlaxGPTJForCausalLM) if is_flax_available() else ()
__lowercase: List[str] = (FlaxGPTJForCausalLM,) if is_flax_available() else ()
def lowerCAmelCase ( self : Tuple ) ->List[str]:
"""simple docstring"""
snake_case_ = FlaxGPTJModelTester(self )
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] ) ->Any:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward_with_attn_mask(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
@tooslow
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = GPTaTokenizer.from_pretrained("""gpt2""" , pad_token="""<|endoftext|>""" , padding_side="""left""" )
snake_case_ = tokenizer(["""Hello this is a long string""", """Hey"""] , return_tensors="""np""" , padding=UpperCAmelCase_ , truncation=UpperCAmelCase_ )
snake_case_ = FlaxGPTJForCausalLM.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = False
snake_case_ = model.config.eos_token_id
snake_case_ = jax.jit(model.generate )
snake_case_ = jit_generate(
inputs["""input_ids"""] , attention_mask=inputs["""attention_mask"""] , pad_token_id=tokenizer.pad_token_id ).sequences
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = [
"""Hello this is a long string of text.\n\nI'm trying to get the text of the""",
"""Hey, I'm a little late to the party. I'm going to""",
]
self.assertListEqual(UpperCAmelCase_ , UpperCAmelCase_ )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = convert_pytorch_state_dict_to_flax(pt_model.state_dict() , UpperCAmelCase_ )
snake_case_ = fx_state
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
pt_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = model_class.from_pretrained(UpperCAmelCase_ , from_pt=UpperCAmelCase_ )
snake_case_ = fx_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output_loaded, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output_loaded[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = load_flax_weights_in_pytorch_model(UpperCAmelCase_ , fx_model.params )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
# make sure weights are tied in PyTorch
pt_model.tie_weights()
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
fx_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = pt_model_class.from_pretrained(UpperCAmelCase_ , from_flax=UpperCAmelCase_ )
with torch.no_grad():
snake_case_ = pt_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@tooslow
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
import os
import unittest
from huggingface_hub.utils import are_progress_bars_disabled
import transformers.models.bart.tokenization_bart
from transformers import logging
from transformers.testing_utils import CaptureLogger, mockenv, mockenv_context
from transformers.utils.logging import disable_progress_bar, enable_progress_bar
class UpperCAmelCase ( unittest.TestCase ):
'''simple docstring'''
def UpperCamelCase_ ( self : int ):
__A = logging.get_logger()
# the current default level is logging.WARNING
__A = logging.get_verbosity()
logging.set_verbosity_error()
self.assertEqual(logger.getEffectiveLevel() ,logging.get_verbosity() )
logging.set_verbosity_warning()
self.assertEqual(logger.getEffectiveLevel() ,logging.get_verbosity() )
logging.set_verbosity_info()
self.assertEqual(logger.getEffectiveLevel() ,logging.get_verbosity() )
logging.set_verbosity_debug()
self.assertEqual(logger.getEffectiveLevel() ,logging.get_verbosity() )
# restore to the original level
logging.set_verbosity(A )
def UpperCamelCase_ ( self : Dict ):
__A = logging.get_verbosity()
__A = logging.get_logger("transformers.models.bart.tokenization_bart" )
__A = "Testing 1, 2, 3"
# should be able to log warnings (if default settings weren't overridden by `pytest --log-level-all`)
if level_origin <= logging.WARNING:
with CaptureLogger(A ) as cl:
logger.warning(A )
self.assertEqual(cl.out ,msg + "\n" )
# this is setting the level for all of `transformers.*` loggers
logging.set_verbosity_error()
# should not be able to log warnings
with CaptureLogger(A ) as cl:
logger.warning(A )
self.assertEqual(cl.out ,"" )
# should be able to log warnings again
logging.set_verbosity_warning()
with CaptureLogger(A ) as cl:
logger.warning(A )
self.assertEqual(cl.out ,msg + "\n" )
# restore to the original level
logging.set_verbosity(A )
@mockenv(TRANSFORMERS_VERBOSITY="error" )
def UpperCamelCase_ ( self : Optional[Any] ):
# reset for the env var to take effect, next time some logger call is made
transformers.utils.logging._reset_library_root_logger()
# this action activates the env var
__A = logging.get_logger("transformers.models.bart.tokenization_bart" )
__A = os.getenv("TRANSFORMERS_VERBOSITY" ,A )
__A = logging.log_levels[env_level_str]
__A = logging.get_verbosity()
self.assertEqual(
A ,A ,f'''TRANSFORMERS_VERBOSITY={env_level_str}/{env_level}, but internal verbosity is {current_level}''' ,)
# restore to the original level
__A = ""
transformers.utils.logging._reset_library_root_logger()
@mockenv(TRANSFORMERS_VERBOSITY="super-error" )
def UpperCamelCase_ ( self : Tuple ):
# reset for the env var to take effect, next time some logger call is made
transformers.utils.logging._reset_library_root_logger()
__A = logging.logging.getLogger()
with CaptureLogger(A ) as cl:
# this action activates the env var
logging.get_logger("transformers.models.bart.tokenization_bart" )
self.assertIn("Unknown option TRANSFORMERS_VERBOSITY=super-error" ,cl.out )
# no need to restore as nothing was changed
def UpperCamelCase_ ( self : int ):
# testing `logger.warning_advice()`
transformers.utils.logging._reset_library_root_logger()
__A = logging.get_logger("transformers.models.bart.tokenization_bart" )
__A = "Testing 1, 2, 3"
with mockenv_context(TRANSFORMERS_NO_ADVISORY_WARNINGS="1" ):
# nothing should be logged as env var disables this method
with CaptureLogger(A ) as cl:
logger.warning_advice(A )
self.assertEqual(cl.out ,"" )
with mockenv_context(TRANSFORMERS_NO_ADVISORY_WARNINGS="" ):
# should log normally as TRANSFORMERS_NO_ADVISORY_WARNINGS is unset
with CaptureLogger(A ) as cl:
logger.warning_advice(A )
self.assertEqual(cl.out ,msg + "\n" )
def UpperCAmelCase ( ) -> List[str]:
"""simple docstring"""
disable_progress_bar()
assert are_progress_bars_disabled()
enable_progress_bar()
assert not are_progress_bars_disabled()
| 15 |
"""simple docstring"""
import copy
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ..auto.configuration_auto import CONFIG_MAPPING
__SCREAMING_SNAKE_CASE : Any = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """upernet"""
def __init__( self : str , UpperCAmelCase_ : List[str]=None , UpperCAmelCase_ : str=512 , UpperCAmelCase_ : int=0.02 , UpperCAmelCase_ : Optional[Any]=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Tuple=0.4 , UpperCAmelCase_ : Tuple=384 , UpperCAmelCase_ : Union[str, Any]=256 , UpperCAmelCase_ : str=1 , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : Dict , ) ->Union[str, Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
if backbone_config is None:
logger.info("""`backbone_config` is `None`. Initializing the config with the default `ResNet` backbone.""" )
snake_case_ = CONFIG_MAPPING["""resnet"""](out_features=["""stage1""", """stage2""", """stage3""", """stage4"""] )
elif isinstance(UpperCAmelCase_ , UpperCAmelCase_ ):
snake_case_ = backbone_config.get("""model_type""" )
snake_case_ = CONFIG_MAPPING[backbone_model_type]
snake_case_ = config_class.from_dict(UpperCAmelCase_ )
snake_case_ = backbone_config
snake_case_ = hidden_size
snake_case_ = initializer_range
snake_case_ = pool_scales
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_in_channels
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = loss_ignore_index
def lowerCAmelCase ( self : str ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = copy.deepcopy(self.__dict__ )
snake_case_ = self.backbone_config.to_dict()
snake_case_ = self.__class__.model_type
return output
| 347 | 0 |
"""simple docstring"""
import inspect
import os
import unittest
import torch
import accelerate
from accelerate import Accelerator
from accelerate.test_utils import execute_subprocess_async, require_multi_gpu
from accelerate.utils import patch_environment
class __A ( unittest.TestCase ):
'''simple docstring'''
def UpperCAmelCase ( self : List[Any] ) -> Tuple:
"""simple docstring"""
lowercase__ : List[Any] = inspect.getfile(accelerate.test_utils )
lowercase__ : Tuple = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ['''scripts''', '''test_script.py'''] )
lowercase__ : Tuple = os.path.sep.join(
mod_file.split(os.path.sep )[:-1] + ['''scripts''', '''test_distributed_data_loop.py'''] )
lowercase__ : Any = os.path.sep.join(mod_file.split(os.path.sep )[:-1] + ['''scripts''', '''test_ops.py'''] )
@require_multi_gpu
def UpperCAmelCase ( self : Dict ) -> Any:
"""simple docstring"""
print(f"""Found {torch.cuda.device_count()} devices.""" )
lowercase__ : Any = ['''torchrun''', f"""--nproc_per_node={torch.cuda.device_count()}""", self.test_file_path]
with patch_environment(omp_num_threads=1 ):
execute_subprocess_async(_snake_case ,env=os.environ.copy() )
@require_multi_gpu
def UpperCAmelCase ( self : str ) -> List[Any]:
"""simple docstring"""
print(f"""Found {torch.cuda.device_count()} devices.""" )
lowercase__ : Dict = ['''torchrun''', f"""--nproc_per_node={torch.cuda.device_count()}""", self.operation_file_path]
print(f"""Command: {cmd}""" )
with patch_environment(omp_num_threads=1 ):
execute_subprocess_async(_snake_case ,env=os.environ.copy() )
@require_multi_gpu
def UpperCAmelCase ( self : Optional[int] ) -> str:
"""simple docstring"""
lowercase__ : Any = ['''torchrun''', f"""--nproc_per_node={torch.cuda.device_count()}""", inspect.getfile(self.__class__ )]
with patch_environment(omp_num_threads=1 ):
execute_subprocess_async(_snake_case ,env=os.environ.copy() )
@require_multi_gpu
def UpperCAmelCase ( self : Dict ) -> Optional[Any]:
"""simple docstring"""
print(f"""Found {torch.cuda.device_count()} devices, using 2 devices only""" )
lowercase__ : str = ['''torchrun''', f"""--nproc_per_node={torch.cuda.device_count()}""", self.data_loop_file_path]
with patch_environment(omp_num_threads=1 ,cuda_visible_devices='''0,1''' ):
execute_subprocess_async(_snake_case ,env=os.environ.copy() )
if __name__ == "__main__":
lowerCAmelCase_ = Accelerator()
lowerCAmelCase_ = (accelerator.state.process_index + 2, 10)
lowerCAmelCase_ = torch.randint(0, 10, shape).to(accelerator.device)
lowerCAmelCase_ = ''
lowerCAmelCase_ = accelerator.pad_across_processes(tensor)
if tensora.shape[0] != accelerator.state.num_processes + 1:
error_msg += F"Found shape {tensora.shape} but should have {accelerator.state.num_processes + 1} at dim 0."
if not torch.equal(tensora[: accelerator.state.process_index + 2], tensor):
error_msg += "Tensors have different values."
if not torch.all(tensora[accelerator.state.process_index + 2 :] == 0):
error_msg += "Padding was not done with the right value (0)."
lowerCAmelCase_ = accelerator.pad_across_processes(tensor, pad_first=True)
if tensora.shape[0] != accelerator.state.num_processes + 1:
error_msg += F"Found shape {tensora.shape} but should have {accelerator.state.num_processes + 1} at dim 0."
lowerCAmelCase_ = accelerator.state.num_processes - accelerator.state.process_index - 1
if not torch.equal(tensora[index:], tensor):
error_msg += "Tensors have different values."
if not torch.all(tensora[:index] == 0):
error_msg += "Padding was not done with the right value (0)."
# Raise error at the end to make sure we don't stop at the first failure.
if len(error_msg) > 0:
raise ValueError(error_msg)
| 16 |
"""simple docstring"""
import os
import shutil
import tempfile
import unittest
import numpy as np
from transformers import AutoTokenizer, BarkProcessor
from transformers.testing_utils import require_torch, slow
@require_torch
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = """ylacombe/bark-small"""
snake_case_ = tempfile.mkdtemp()
snake_case_ = """en_speaker_1"""
snake_case_ = """This is a test string"""
snake_case_ = """speaker_embeddings_path.json"""
snake_case_ = """speaker_embeddings"""
def lowerCAmelCase ( self : List[str] , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
return AutoTokenizer.from_pretrained(self.checkpoint , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
shutil.rmtree(self.tmpdirname )
def lowerCAmelCase ( self : List[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
snake_case_ = BarkProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
@slow
def lowerCAmelCase ( self : Dict ) ->int:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
processor.save_pretrained(
self.tmpdirname , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , speaker_embeddings_directory=self.speaker_embeddings_directory , )
snake_case_ = self.get_tokenizer(bos_token="""(BOS)""" , eos_token="""(EOS)""" )
snake_case_ = BarkProcessor.from_pretrained(
self.tmpdirname , self.speaker_embeddings_dict_path , bos_token="""(BOS)""" , eos_token="""(EOS)""" , )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
def lowerCAmelCase ( self : Optional[Any] ) ->Any:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
snake_case_ = 35
snake_case_ = 2
snake_case_ = 8
snake_case_ = {
"""semantic_prompt""": np.ones(UpperCAmelCase_ ),
"""coarse_prompt""": np.ones((nb_codebooks_coarse, seq_len) ),
"""fine_prompt""": np.ones((nb_codebooks_total, seq_len) ),
}
# test providing already loaded voice_preset
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from npz file
snake_case_ = os.path.join(self.tmpdirname , """file.npz""" )
np.savez(UpperCAmelCase_ , **UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from the hub
snake_case_ = processor(text=self.input_string , voice_preset=self.voice_preset )
def lowerCAmelCase ( self : Tuple ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string )
snake_case_ = tokenizer(
self.input_string , padding="""max_length""" , max_length=256 , add_special_tokens=UpperCAmelCase_ , return_attention_mask=UpperCAmelCase_ , return_token_type_ids=UpperCAmelCase_ , )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key].squeeze().tolist() )
| 347 | 0 |
"""simple docstring"""
def _A ( UpperCamelCase_ : Any) -> List[str]:
'''simple docstring'''
__lowercase ,__lowercase = [], []
while len(UpperCamelCase_) > 1:
__lowercase ,__lowercase = min(UpperCamelCase_), max(UpperCamelCase_)
start.append(UpperCamelCase_)
end.append(UpperCamelCase_)
collection.remove(UpperCamelCase_)
collection.remove(UpperCamelCase_)
end.reverse()
return start + collection + end
if __name__ == "__main__":
_a = input('Enter numbers separated by a comma:\n').strip()
_a = [int(item) for item in user_input.split(',')]
print(*merge_sort(unsorted), sep=',')
| 17 |
"""simple docstring"""
import argparse
import json
import os
import sys
import tempfile
import unittest
from argparse import Namespace
from dataclasses import dataclass, field
from enum import Enum
from pathlib import Path
from typing import List, Literal, Optional
import yaml
from transformers import HfArgumentParser, TrainingArguments
from transformers.hf_argparser import make_choice_type_function, string_to_bool
# Since Python 3.10, we can use the builtin `|` operator for Union types
# See PEP 604: https://peps.python.org/pep-0604
__SCREAMING_SNAKE_CASE : int = sys.version_info >= (3, 10)
def _a ( _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
return field(default_factory=lambda: default , metadata=_SCREAMING_SNAKE_CASE )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: float
__lowercase: str
__lowercase: bool
@dataclass
class __A :
'''simple docstring'''
__lowercase: int = 42
__lowercase: str = field(default="""toto""" , metadata={"""help""": """help message"""})
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: Optional[bool] = None
class __A (snake_case__):
'''simple docstring'''
__lowercase: str = """titi"""
__lowercase: Any = """toto"""
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """titi"""
__lowercase: Optional[Any] = """toto"""
__lowercase: List[Any] = 42
@dataclass
class __A :
'''simple docstring'''
__lowercase: BasicEnum = "toto"
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
snake_case_ = BasicEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: MixedTypeEnum = "toto"
def lowerCAmelCase ( self : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = MixedTypeEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: Optional[int] = None
__lowercase: Optional[float] = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: Optional[str] = None
__lowercase: Optional[List[str]] = list_field(default=[])
__lowercase: Optional[List[int]] = list_field(default=[])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = list_field(default=[])
__lowercase: List[int] = list_field(default=[1, 2, 3])
__lowercase: List[str] = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
__lowercase: List[float] = list_field(default=[0.1, 0.2, 0.3])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = field()
__lowercase: str = field()
__lowercase: BasicEnum = field()
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
snake_case_ = BasicEnum(self.required_enum )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: "BasicEnum" = field()
__lowercase: "Optional[bool]" = None
__lowercase: "str" = field(default="""toto""" , metadata={"""help""": """help message"""})
__lowercase: "List[str]" = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
if is_python_no_less_than_3_10:
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: bool | None = None
@dataclass
class __A :
'''simple docstring'''
__lowercase: int | None = None
__lowercase: float | None = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: str | None = None
__lowercase: list[str] | None = list_field(default=[])
__lowercase: list[int] | None = list_field(default=[])
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : argparse.ArgumentParser , UpperCAmelCase_ : argparse.ArgumentParser ) ->Optional[int]:
"""simple docstring"""
self.assertEqual(len(a._actions ) , len(b._actions ) )
for x, y in zip(a._actions , b._actions ):
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
# Choices with mixed type have custom function as "type"
# So we need to compare results directly for equality
if xx.get("""choices""" , UpperCAmelCase_ ) and yy.get("""choices""" , UpperCAmelCase_ ):
for expected_choice in yy["choices"] + xx["choices"]:
self.assertEqual(xx["""type"""](UpperCAmelCase_ ) , yy["""type"""](UpperCAmelCase_ ) )
del xx["type"], yy["type"]
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--bar""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--flag""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = ["""--foo""", """1""", """--baz""", """quux""", """--bar""", """0.5"""]
((snake_case_) , ) = parser.parse_args_into_dataclasses(UpperCAmelCase_ , look_for_args_file=UpperCAmelCase_ )
self.assertFalse(example.flag )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=42 , type=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
# A boolean no_* argument always has to come after its "default: True" regular counter-part
# and its default must be set to False
expected.add_argument("""--no_baz""" , action="""store_false""" , default=UpperCAmelCase_ , dest="""baz""" )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
snake_case_ = [WithDefaultBoolExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--no_baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """True""", """--baz""", """True""", """--opt""", """True"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """False""", """--baz""", """False""", """--opt""", """False"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
def lowerCAmelCase ( self : int ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=["""titi""", """toto""", 42] , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args_into_dataclasses([] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.toto )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """titi"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.titi )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """42"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.fourtytwo )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
@dataclass
class __A :
'''simple docstring'''
__lowercase: Literal["titi", "toto", 42] = "toto"
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=("""titi""", """toto""", 42) , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo_int""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--bar_int""" , nargs="""+""" , default=[1, 2, 3] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_float""" , nargs="""+""" , default=[0.1, 0.2, 0.3] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(
UpperCAmelCase_ , Namespace(foo_int=[] , bar_int=[1, 2, 3] , foo_str=["""Hallo""", """Bonjour""", """Hello"""] , foo_float=[0.1, 0.2, 0.3] ) , )
snake_case_ = parser.parse_args("""--foo_int 1 --bar_int 2 3 --foo_str a b c --foo_float 0.1 0.7""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo_int=[1] , bar_int=[2, 3] , foo_str=["""a""", """b""", """c"""] , foo_float=[0.1, 0.7] ) )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--bar""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--baz""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--ces""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--des""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
snake_case_ = [OptionalExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , bar=UpperCAmelCase_ , baz=UpperCAmelCase_ , ces=[] , des=[] ) )
snake_case_ = parser.parse_args("""--foo 12 --bar 3.14 --baz 42 --ces a b c --des 1 2 3""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=12 , bar=3.14 , baz="""42""" , ces=["""a""", """b""", """c"""] , des=[1, 2, 3] ) )
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--required_list""" , nargs="""+""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--required_str""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
snake_case_ = parser.parse_dict(UpperCAmelCase_ )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
"""extra""": 42,
}
self.assertRaises(UpperCAmelCase_ , parser.parse_dict , UpperCAmelCase_ , allow_extra_keys=UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_json""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.json""" , """w+""" ) as f:
json.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.json""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_yaml""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.yaml""" , """w+""" ) as f:
yaml.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.yaml""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
import argparse
import json
from typing import List
from ltp import LTP
from transformers import BertTokenizer
def _snake_case ( lowerCAmelCase : Dict ):
"""simple docstring"""
if (
(cp >= 0X4_E_0_0 and cp <= 0X9_F_F_F)
or (cp >= 0X3_4_0_0 and cp <= 0X4_D_B_F) #
or (cp >= 0X2_0_0_0_0 and cp <= 0X2_A_6_D_F) #
or (cp >= 0X2_A_7_0_0 and cp <= 0X2_B_7_3_F) #
or (cp >= 0X2_B_7_4_0 and cp <= 0X2_B_8_1_F) #
or (cp >= 0X2_B_8_2_0 and cp <= 0X2_C_E_A_F) #
or (cp >= 0XF_9_0_0 and cp <= 0XF_A_F_F)
or (cp >= 0X2_F_8_0_0 and cp <= 0X2_F_A_1_F) #
): #
return True
return False
def _snake_case ( lowerCAmelCase : str ):
"""simple docstring"""
for char in word:
SCREAMING_SNAKE_CASE_ : Any = ord(lowerCAmelCase )
if not _is_chinese_char(lowerCAmelCase ):
return 0
return 1
def _snake_case ( lowerCAmelCase : List[str] ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : str = set()
for token in tokens:
SCREAMING_SNAKE_CASE_ : int = len(lowerCAmelCase ) > 1 and is_chinese(lowerCAmelCase )
if chinese_word:
word_set.add(lowerCAmelCase )
SCREAMING_SNAKE_CASE_ : Union[str, Any] = list(lowerCAmelCase )
return word_list
def _snake_case ( lowerCAmelCase : List[str] , lowerCAmelCase : set() ):
"""simple docstring"""
if not chinese_word_set:
return bert_tokens
SCREAMING_SNAKE_CASE_ : str = max([len(lowerCAmelCase ) for w in chinese_word_set] )
SCREAMING_SNAKE_CASE_ : int = bert_tokens
SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ : Any = 0, len(lowerCAmelCase )
while start < end:
SCREAMING_SNAKE_CASE_ : Optional[int] = True
if is_chinese(bert_word[start] ):
SCREAMING_SNAKE_CASE_ : List[str] = min(end - start , lowerCAmelCase )
for i in range(lowerCAmelCase , 1 , -1 ):
SCREAMING_SNAKE_CASE_ : Tuple = "".join(bert_word[start : start + i] )
if whole_word in chinese_word_set:
for j in range(start + 1 , start + i ):
SCREAMING_SNAKE_CASE_ : Optional[int] = "##" + bert_word[j]
SCREAMING_SNAKE_CASE_ : int = start + i
SCREAMING_SNAKE_CASE_ : Optional[Any] = False
break
if single_word:
start += 1
return bert_word
def _snake_case ( lowerCAmelCase : List[str] , lowerCAmelCase : LTP , lowerCAmelCase : BertTokenizer ):
"""simple docstring"""
SCREAMING_SNAKE_CASE_ : Union[str, Any] = []
for i in range(0 , len(lowerCAmelCase ) , 1_0_0 ):
SCREAMING_SNAKE_CASE_ : Optional[int] = ltp_tokenizer.seg(lines[i : i + 1_0_0] )[0]
SCREAMING_SNAKE_CASE_ : List[Any] = [get_chinese_word(lowerCAmelCase ) for r in res]
ltp_res.extend(lowerCAmelCase )
assert len(lowerCAmelCase ) == len(lowerCAmelCase )
SCREAMING_SNAKE_CASE_ : Union[str, Any] = []
for i in range(0 , len(lowerCAmelCase ) , 1_0_0 ):
SCREAMING_SNAKE_CASE_ : str = bert_tokenizer(lines[i : i + 1_0_0] , add_special_tokens=lowerCAmelCase , truncation=lowerCAmelCase , max_length=5_1_2 )
bert_res.extend(res["input_ids"] )
assert len(lowerCAmelCase ) == len(lowerCAmelCase )
SCREAMING_SNAKE_CASE_ : int = []
for input_ids, chinese_word in zip(lowerCAmelCase , lowerCAmelCase ):
SCREAMING_SNAKE_CASE_ : str = []
for id in input_ids:
SCREAMING_SNAKE_CASE_ : Tuple = bert_tokenizer._convert_id_to_token(lowerCAmelCase )
input_tokens.append(lowerCAmelCase )
SCREAMING_SNAKE_CASE_ : Any = add_sub_symbol(lowerCAmelCase , lowerCAmelCase )
SCREAMING_SNAKE_CASE_ : Tuple = []
# We only save pos of chinese subwords start with ##, which mean is part of a whole word.
for i, token in enumerate(lowerCAmelCase ):
if token[:2] == "##":
SCREAMING_SNAKE_CASE_ : List[Any] = token[2:]
# save chinese tokens' pos
if len(lowerCAmelCase ) == 1 and _is_chinese_char(ord(lowerCAmelCase ) ):
ref_id.append(lowerCAmelCase )
ref_ids.append(lowerCAmelCase )
assert len(lowerCAmelCase ) == len(lowerCAmelCase )
return ref_ids
def _snake_case ( lowerCAmelCase : Optional[int] ):
"""simple docstring"""
with open(args.file_name , "r" , encoding="utf-8" ) as f:
SCREAMING_SNAKE_CASE_ : List[Any] = f.readlines()
SCREAMING_SNAKE_CASE_ : int = [line.strip() for line in data if len(lowerCAmelCase ) > 0 and not line.isspace()] # avoid delimiter like '\u2029'
SCREAMING_SNAKE_CASE_ : Union[str, Any] = LTP(args.ltp ) # faster in GPU device
SCREAMING_SNAKE_CASE_ : Tuple = BertTokenizer.from_pretrained(args.bert )
SCREAMING_SNAKE_CASE_ : List[str] = prepare_ref(lowerCAmelCase , lowerCAmelCase , lowerCAmelCase )
with open(args.save_path , "w" , encoding="utf-8" ) as f:
SCREAMING_SNAKE_CASE_ : Optional[int] = [json.dumps(lowerCAmelCase ) + "\n" for ref in ref_ids]
f.writelines(lowerCAmelCase )
if __name__ == "__main__":
__lowerCamelCase : Any = argparse.ArgumentParser(description='''prepare_chinese_ref''')
parser.add_argument(
'''--file_name''',
type=str,
default='''./resources/chinese-demo.txt''',
help='''file need process, same as training data in lm''',
)
parser.add_argument(
'''--ltp''', type=str, default='''./resources/ltp''', help='''resources for LTP tokenizer, usually a path'''
)
parser.add_argument('''--bert''', type=str, default='''./resources/robert''', help='''resources for Bert tokenizer''')
parser.add_argument('''--save_path''', type=str, default='''./resources/ref.txt''', help='''path to save res''')
__lowerCamelCase : int = parser.parse_args()
main(args)
| 18 |
"""simple docstring"""
import logging
import os
from typing import Dict, List, Optional, Union
import torch
import torch.nn as nn
from accelerate.utils.imports import (
is_abit_bnb_available,
is_abit_bnb_available,
is_bnb_available,
)
from ..big_modeling import dispatch_model, init_empty_weights
from .dataclasses import BnbQuantizationConfig
from .modeling import (
find_tied_parameters,
get_balanced_memory,
infer_auto_device_map,
load_checkpoint_in_model,
offload_weight,
set_module_tensor_to_device,
)
if is_bnb_available():
import bitsandbytes as bnb
from copy import deepcopy
__SCREAMING_SNAKE_CASE : Any = logging.getLogger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = False , ) -> Optional[Any]:
snake_case_ = bnb_quantization_config.load_in_abit
snake_case_ = bnb_quantization_config.load_in_abit
if load_in_abit and not is_abit_bnb_available():
raise ImportError(
"""You have a version of `bitsandbytes` that is not compatible with 8bit quantization,"""
""" make sure you have the latest version of `bitsandbytes` installed.""" )
if load_in_abit and not is_abit_bnb_available():
raise ValueError(
"""You have a version of `bitsandbytes` that is not compatible with 4bit quantization,"""
"""make sure you have the latest version of `bitsandbytes` installed.""" )
snake_case_ = []
# custom device map
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) and len(device_map.keys() ) > 1:
snake_case_ = [key for key, value in device_map.items() if value in ["""disk""", """cpu"""]]
# We keep some modules such as the lm_head in their original dtype for numerical stability reasons
if bnb_quantization_config.skip_modules is None:
snake_case_ = get_keys_to_not_convert(_SCREAMING_SNAKE_CASE )
# add cpu modules to skip modules only for 4-bit modules
if load_in_abit:
bnb_quantization_config.skip_modules.extend(_SCREAMING_SNAKE_CASE )
snake_case_ = bnb_quantization_config.skip_modules
# We add the modules we want to keep in full precision
if bnb_quantization_config.keep_in_fpaa_modules is None:
snake_case_ = []
snake_case_ = bnb_quantization_config.keep_in_fpaa_modules
modules_to_not_convert.extend(_SCREAMING_SNAKE_CASE )
# compatibility with peft
snake_case_ = load_in_abit
snake_case_ = load_in_abit
snake_case_ = get_parameter_device(_SCREAMING_SNAKE_CASE )
if model_device.type != "meta":
# quantization of an already loaded model
logger.warning(
"""It is not recommended to quantize a loaded model. """
"""The model should be instantiated under the `init_empty_weights` context manager.""" )
snake_case_ = replace_with_bnb_layers(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
# convert param to the right dtype
snake_case_ = bnb_quantization_config.torch_dtype
for name, param in model.state_dict().items():
if any(module_to_keep_in_fpaa in name for module_to_keep_in_fpaa in keep_in_fpaa_modules ):
param.to(torch.floataa )
if param.dtype != torch.floataa:
snake_case_ = name.replace(""".weight""" , """""" ).replace(""".bias""" , """""" )
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if param is not None:
param.to(torch.floataa )
elif torch.is_floating_point(_SCREAMING_SNAKE_CASE ):
param.to(_SCREAMING_SNAKE_CASE )
if model_device.type == "cuda":
# move everything to cpu in the first place because we can't do quantization if the weights are already on cuda
model.cuda(torch.cuda.current_device() )
torch.cuda.empty_cache()
elif torch.cuda.is_available():
model.to(torch.cuda.current_device() )
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info(
f"""The model device type is {model_device.type}. However, cuda is needed for quantization."""
"""We move the model to cuda.""" )
return model
elif weights_location is None:
raise RuntimeError(
f"""`weights_location` needs to be the folder path containing the weights of the model, but we found {weights_location} """ )
else:
with init_empty_weights():
snake_case_ = replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
snake_case_ = get_quantized_model_device_map(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , max_memory=_SCREAMING_SNAKE_CASE , no_split_module_classes=_SCREAMING_SNAKE_CASE , )
if offload_state_dict is None and device_map is not None and "disk" in device_map.values():
snake_case_ = True
snake_case_ = any(x in list(device_map.values() ) for x in ["""cpu""", """disk"""] )
load_checkpoint_in_model(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , dtype=bnb_quantization_config.torch_dtype , offload_folder=_SCREAMING_SNAKE_CASE , offload_state_dict=_SCREAMING_SNAKE_CASE , keep_in_fpaa_modules=bnb_quantization_config.keep_in_fpaa_modules , offload_abit_bnb=load_in_abit and offload , )
return dispatch_model(_SCREAMING_SNAKE_CASE , device_map=_SCREAMING_SNAKE_CASE , offload_dir=_SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if device_map is None:
if torch.cuda.is_available():
snake_case_ = {"""""": torch.cuda.current_device()}
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info("""The device_map was not initialized.""" """Setting device_map to `{'':torch.cuda.current_device()}`.""" )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
if device_map not in ["auto", "balanced", "balanced_low_0", "sequential"]:
raise ValueError(
"""If passing a string for `device_map`, please choose 'auto', 'balanced', 'balanced_low_0' or """
"""'sequential'.""" )
snake_case_ = {}
special_dtypes.update(
{
name: bnb_quantization_config.torch_dtype
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.skip_modules )
} )
special_dtypes.update(
{
name: torch.floataa
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.keep_in_fpaa_modules )
} )
snake_case_ = {}
snake_case_ = special_dtypes
snake_case_ = no_split_module_classes
snake_case_ = bnb_quantization_config.target_dtype
# get max_memory for each device.
if device_map != "sequential":
snake_case_ = get_balanced_memory(
_SCREAMING_SNAKE_CASE , low_zero=(device_map == """balanced_low_0""") , max_memory=_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE , )
snake_case_ = max_memory
snake_case_ = infer_auto_device_map(_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
# check if don't have any quantized module on the cpu
snake_case_ = bnb_quantization_config.skip_modules + bnb_quantization_config.keep_in_fpaa_modules
snake_case_ = {
key: device_map[key] for key in device_map.keys() if key not in modules_not_to_convert
}
for device in ["cpu", "disk"]:
if device in device_map_without_some_modules.values():
if bnb_quantization_config.load_in_abit:
raise ValueError(
"""
Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit
the quantized model. If you want to dispatch the model on the CPU or the disk while keeping
these modules in `torch_dtype`, you need to pass a custom `device_map` to
`load_and_quantize_model`. Check
https://huggingface.co/docs/accelerate/main/en/usage_guides/quantization#offload-modules-to-cpu-and-disk
for more details.
""" )
else:
logger.info(
"""Some modules are are offloaded to the CPU or the disk. Note that these modules will be converted to 8-bit""" )
del device_map_without_some_modules
return device_map
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if modules_to_not_convert is None:
snake_case_ = []
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not has_been_replaced:
logger.warning(
"""You are loading your model in 8bit or 4bit but no linear modules were found in your model."""
""" this can happen for some architectures such as gpt2 that uses Conv1D instead of Linear layers."""
""" Please double check your model architecture, or submit an issue on github if you think this is"""
""" a bug.""" )
return model
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , ) -> List[Any]:
snake_case_ = False
for name, module in model.named_children():
if current_key_name is None:
snake_case_ = []
current_key_name.append(_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , nn.Linear ) and name not in modules_to_not_convert:
# Check if the current key is not in the `modules_to_not_convert`
snake_case_ = """.""".join(_SCREAMING_SNAKE_CASE )
snake_case_ = True
for key in modules_to_not_convert:
if (
(key in current_key_name_str) and (key + "." in current_key_name_str)
) or key == current_key_name_str:
snake_case_ = False
break
if proceed:
# Load bnb module with empty weight and replace ``nn.Linear` module
if bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.LinearabitLt(
module.in_features , module.out_features , module.bias is not None , has_fpaa_weights=_SCREAMING_SNAKE_CASE , threshold=bnb_quantization_config.llm_inta_threshold , )
elif bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.Linearabit(
module.in_features , module.out_features , module.bias is not None , bnb_quantization_config.bnb_abit_compute_dtype , compress_statistics=bnb_quantization_config.bnb_abit_use_double_quant , quant_type=bnb_quantization_config.bnb_abit_quant_type , )
else:
raise ValueError("""load_in_8bit and load_in_4bit can't be both False""" )
snake_case_ = module.weight.data
if module.bias is not None:
snake_case_ = module.bias.data
bnb_module.requires_grad_(_SCREAMING_SNAKE_CASE )
setattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = True
if len(list(module.children() ) ) > 0:
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = has_been_replaced | _has_been_replaced
# Remove the last key for recursion
current_key_name.pop(-1 )
return model, has_been_replaced
def _a ( _SCREAMING_SNAKE_CASE ) -> Any:
# Create a copy of the model
with init_empty_weights():
snake_case_ = deepcopy(_SCREAMING_SNAKE_CASE ) # this has 0 cost since it is done inside `init_empty_weights` context manager`
snake_case_ = find_tied_parameters(_SCREAMING_SNAKE_CASE )
# For compatibility with Accelerate < 0.18
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
snake_case_ = sum(list(tied_params.values() ) , [] ) + list(tied_params.keys() )
else:
snake_case_ = sum(_SCREAMING_SNAKE_CASE , [] )
snake_case_ = len(_SCREAMING_SNAKE_CASE ) > 0
# Check if it is a base model
snake_case_ = False
if hasattr(_SCREAMING_SNAKE_CASE , """base_model_prefix""" ):
snake_case_ = not hasattr(_SCREAMING_SNAKE_CASE , model.base_model_prefix )
# Ignore this for base models (BertModel, GPT2Model, etc.)
if (not has_tied_params) and is_base_model:
return []
# otherwise they have an attached head
snake_case_ = list(model.named_children() )
snake_case_ = [list_modules[-1][0]]
# add last module together with tied weights
snake_case_ = set(_SCREAMING_SNAKE_CASE ) - set(_SCREAMING_SNAKE_CASE )
snake_case_ = list(set(_SCREAMING_SNAKE_CASE ) ) + list(_SCREAMING_SNAKE_CASE )
# remove ".weight" from the keys
snake_case_ = [""".weight""", """.bias"""]
snake_case_ = []
for name in list_untouched:
for name_to_remove in names_to_remove:
if name_to_remove in name:
snake_case_ = name.replace(_SCREAMING_SNAKE_CASE , """""" )
filtered_module_names.append(_SCREAMING_SNAKE_CASE )
return filtered_module_names
def _a ( _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for m in model.modules():
if isinstance(_SCREAMING_SNAKE_CASE , bnb.nn.Linearabit ):
return True
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]:
return next(parameter.parameters() ).device
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
# if it is not quantized, we quantize and offload the quantized weights and the SCB stats
if fpaa_statistics is None:
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , 0 , dtype=_SCREAMING_SNAKE_CASE , value=_SCREAMING_SNAKE_CASE )
snake_case_ = param_name
snake_case_ = model
if "." in tensor_name:
snake_case_ = tensor_name.split(""".""" )
for split in splits[:-1]:
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if new_module is None:
raise ValueError(f"""{module} has no attribute {split}.""" )
snake_case_ = new_module
snake_case_ = splits[-1]
# offload weights
snake_case_ = False
offload_weight(module._parameters[tensor_name] , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
if hasattr(module._parameters[tensor_name] , """SCB""" ):
offload_weight(
module._parameters[tensor_name].SCB , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE , )
else:
offload_weight(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
offload_weight(_SCREAMING_SNAKE_CASE , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , """meta""" , dtype=_SCREAMING_SNAKE_CASE , value=torch.empty(*param.size() ) )
| 347 | 0 |
import importlib.util
import json
import os
import warnings
from dataclasses import dataclass, field
import torch
from ..training_args import TrainingArguments
from ..utils import cached_property, is_sagemaker_dp_enabled, logging
__A =logging.get_logger(__name__)
def lowerCamelCase_ ( ):
# Get the sagemaker specific mp parameters from smp_options variable.
lowerCamelCase_ = os.getenv("SM_HP_MP_PARAMETERS" , "{}" )
try:
# Parse it and check the field "partitions" is included, it is required for model parallel.
lowerCamelCase_ = json.loads(lowerCamelCase__ )
if "partitions" not in smp_options:
return False
except json.JSONDecodeError:
return False
# Get the sagemaker specific framework parameters from mpi_options variable.
lowerCamelCase_ = os.getenv("SM_FRAMEWORK_PARAMS" , "{}" )
try:
# Parse it and check the field "sagemaker_distributed_dataparallel_enabled".
lowerCamelCase_ = json.loads(lowerCamelCase__ )
if not mpi_options.get("sagemaker_mpi_enabled" , lowerCamelCase__ ):
return False
except json.JSONDecodeError:
return False
# Lastly, check if the `smdistributed` module is present.
return importlib.util.find_spec("smdistributed" ) is not None
if is_sagemaker_model_parallel_available():
import smdistributed.modelparallel.torch as smp
smp.init()
@dataclass
class _SCREAMING_SNAKE_CASE ( snake_case_ ):
lowerCAmelCase__ = field(
default='' , metadata={'help': 'Used by the SageMaker launcher to send mp-specific args. Ignored in SageMakerTrainer'} , )
def SCREAMING_SNAKE_CASE_( self ) -> Tuple:
super().__post_init__()
warnings.warn(
"`SageMakerTrainingArguments` is deprecated and will be removed in v5 of Transformers. You can use "
"`TrainingArguments` instead." , lowercase , )
@cached_property
def SCREAMING_SNAKE_CASE_( self ) -> "torch.device":
logger.info("PyTorch: setting up devices" )
if torch.distributed.is_available() and torch.distributed.is_initialized() and self.local_rank == -1:
logger.warning(
"torch.distributed process group is initialized, but local_rank == -1. "
"In order to use Torch DDP, launch your script with `python -m torch.distributed.launch" )
if self.no_cuda:
lowerCamelCase_ = torch.device("cpu" )
lowerCamelCase_ = 0
elif is_sagemaker_model_parallel_available():
lowerCamelCase_ = smp.local_rank()
lowerCamelCase_ = torch.device("cuda" , lowercase )
lowerCamelCase_ = 1
elif is_sagemaker_dp_enabled():
import smdistributed.dataparallel.torch.torch_smddp # noqa: F401
torch.distributed.init_process_group(backend="smddp" , timeout=self.ddp_timeout_delta )
lowerCamelCase_ = int(os.getenv("SMDATAPARALLEL_LOCAL_RANK" ) )
lowerCamelCase_ = torch.device("cuda" , self.local_rank )
lowerCamelCase_ = 1
elif self.local_rank == -1:
# if n_gpu is > 1 we'll use nn.DataParallel.
# If you only want to use a specific subset of GPUs use `CUDA_VISIBLE_DEVICES=0`
# Explicitly set CUDA to the first (index 0) CUDA device, otherwise `set_device` will
# trigger an error that a device index is missing. Index 0 takes into account the
# GPUs available in the environment, so `CUDA_VISIBLE_DEVICES=1,2` with `cuda:0`
# will use the first GPU in that env, i.e. GPU#1
lowerCamelCase_ = torch.device("cuda:0" if torch.cuda.is_available() else "cpu" )
# Sometimes the line in the postinit has not been run before we end up here, so just checking we're not at
# the default value.
lowerCamelCase_ = torch.cuda.device_count()
else:
# Here, we'll use torch.distributed.
# Initializes the distributed backend which will take care of synchronizing nodes/GPUs
if not torch.distributed.is_initialized():
torch.distributed.init_process_group(backend="nccl" , timeout=self.ddp_timeout_delta )
lowerCamelCase_ = torch.device("cuda" , self.local_rank )
lowerCamelCase_ = 1
if device.type == "cuda":
torch.cuda.set_device(lowercase )
return device
@property
def SCREAMING_SNAKE_CASE_( self ) -> Tuple:
if is_sagemaker_model_parallel_available():
return smp.dp_size()
return super().world_size
@property
def SCREAMING_SNAKE_CASE_( self ) -> List[str]:
return not is_sagemaker_model_parallel_available()
@property
def SCREAMING_SNAKE_CASE_( self ) -> Dict:
return False
| 19 |
"""simple docstring"""
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__SCREAMING_SNAKE_CASE : str = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'microsoft/beit-base-patch16-224-pt22k': (
'https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json'
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = """beit"""
def __init__( self : List[str] , UpperCAmelCase_ : List[Any]=8_192 , UpperCAmelCase_ : Dict=768 , UpperCAmelCase_ : int=12 , UpperCAmelCase_ : Tuple=12 , UpperCAmelCase_ : List[Any]=3_072 , UpperCAmelCase_ : Tuple="gelu" , UpperCAmelCase_ : Dict=0.0 , UpperCAmelCase_ : List[str]=0.0 , UpperCAmelCase_ : Any=0.02 , UpperCAmelCase_ : Optional[Any]=1E-12 , UpperCAmelCase_ : int=224 , UpperCAmelCase_ : Tuple=16 , UpperCAmelCase_ : List[str]=3 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[str]=False , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[Any]=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Any=True , UpperCAmelCase_ : Dict=[3, 5, 7, 11] , UpperCAmelCase_ : Tuple=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : List[Any]=0.4 , UpperCAmelCase_ : Optional[Any]=256 , UpperCAmelCase_ : Optional[Any]=1 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : List[str] , ) ->Optional[Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = initializer_range
snake_case_ = layer_norm_eps
snake_case_ = image_size
snake_case_ = patch_size
snake_case_ = num_channels
snake_case_ = use_mask_token
snake_case_ = use_absolute_position_embeddings
snake_case_ = use_relative_position_bias
snake_case_ = use_shared_relative_position_bias
snake_case_ = layer_scale_init_value
snake_case_ = drop_path_rate
snake_case_ = use_mean_pooling
# decode head attributes (semantic segmentation)
snake_case_ = out_indices
snake_case_ = pool_scales
# auxiliary head attributes (semantic segmentation)
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = semantic_loss_ignore_index
class __A (snake_case__):
'''simple docstring'''
__lowercase: List[Any] = version.parse("""1.11""")
@property
def lowerCAmelCase ( self : Dict ) ->Mapping[str, Mapping[int, str]]:
"""simple docstring"""
return OrderedDict(
[
("""pixel_values""", {0: """batch""", 1: """num_channels""", 2: """height""", 3: """width"""}),
] )
@property
def lowerCAmelCase ( self : Any ) ->float:
"""simple docstring"""
return 1E-4
| 347 | 0 |
from typing import Tuple, Union
from ...modeling_outputs import BackboneOutput
from ...modeling_utils import PreTrainedModel
from ...utils import is_timm_available, is_torch_available, requires_backends
from ...utils.backbone_utils import BackboneMixin
from .configuration_timm_backbone import TimmBackboneConfig
if is_timm_available():
import timm
if is_torch_available():
from torch import Tensor
class __snake_case ( lowerCAmelCase , lowerCAmelCase ):
_a : int= "pixel_values"
_a : List[str]= False
_a : Union[str, Any]= TimmBackboneConfig
def __init__( self ,snake_case ,**snake_case ):
'''simple docstring'''
requires_backends(self ,"""timm""" )
super().__init__(snake_case )
lowercase : int = config
if config.backbone is None:
raise ValueError("""backbone is not set in the config. Please set it to a timm model name.""" )
if config.backbone not in timm.list_models():
raise ValueError(f"backbone {config.backbone} is not supported by timm." )
if hasattr(snake_case ,"""out_features""" ) and config.out_features is not None:
raise ValueError("""out_features is not supported by TimmBackbone. Please use out_indices instead.""" )
lowercase : str = getattr(snake_case ,"""use_pretrained_backbone""" ,snake_case )
if pretrained is None:
raise ValueError("""use_pretrained_backbone is not set in the config. Please set it to True or False.""" )
# We just take the final layer by default. This matches the default for the transformers models.
lowercase : Optional[int] = config.out_indices if getattr(snake_case ,"""out_indices""" ,snake_case ) is not None else (-1,)
lowercase : List[Any] = timm.create_model(
config.backbone ,pretrained=snake_case ,features_only=config.features_only ,in_chans=config.num_channels ,out_indices=snake_case ,**snake_case ,)
# These are used to control the output of the model when called. If output_hidden_states is True, then
# return_layers is modified to include all layers.
lowercase : Optional[int] = self._backbone.return_layers
lowercase : Tuple = {layer["""module"""]: str(snake_case ) for i, layer in enumerate(self._backbone.feature_info.info )}
super()._init_backbone(snake_case )
@classmethod
def _SCREAMING_SNAKE_CASE ( cls ,snake_case ,*snake_case ,**snake_case ):
'''simple docstring'''
requires_backends(cls ,["""vision""", """timm"""] )
from ...models.timm_backbone import TimmBackboneConfig
lowercase : Tuple = kwargs.pop("""config""" ,TimmBackboneConfig() )
lowercase : int = kwargs.pop("""use_timm_backbone""" ,snake_case )
if not use_timm:
raise ValueError("""use_timm_backbone must be True for timm backbones""" )
lowercase : Any = kwargs.pop("""num_channels""" ,config.num_channels )
lowercase : List[Any] = kwargs.pop("""features_only""" ,config.features_only )
lowercase : Any = kwargs.pop("""use_pretrained_backbone""" ,config.use_pretrained_backbone )
lowercase : Dict = kwargs.pop("""out_indices""" ,config.out_indices )
lowercase : Any = TimmBackboneConfig(
backbone=snake_case ,num_channels=snake_case ,features_only=snake_case ,use_pretrained_backbone=snake_case ,out_indices=snake_case ,)
return super()._from_config(snake_case ,**snake_case )
def _SCREAMING_SNAKE_CASE ( self ,snake_case ):
'''simple docstring'''
pass
def _SCREAMING_SNAKE_CASE ( self ,snake_case ,snake_case=None ,snake_case=None ,snake_case=None ,**snake_case ):
'''simple docstring'''
lowercase : List[Any] = return_dict if return_dict is not None else self.config.use_return_dict
lowercase : Union[str, Any] = (
output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
)
lowercase : Any = output_attentions if output_attentions is not None else self.config.output_attentions
if output_attentions:
raise ValueError("""Cannot output attentions for timm backbones at the moment""" )
if output_hidden_states:
# We modify the return layers to include all the stages of the backbone
lowercase : str = self._all_layers
lowercase : Any = self._backbone(snake_case ,**snake_case )
lowercase : List[str] = self._return_layers
lowercase : List[Any] = tuple(hidden_states[i] for i in self.out_indices )
else:
lowercase : List[Any] = self._backbone(snake_case ,**snake_case )
lowercase : Tuple = None
lowercase : Any = tuple(snake_case )
lowercase : Dict = tuple(snake_case ) if hidden_states is not None else None
if not return_dict:
lowercase : Union[str, Any] = (feature_maps,)
if output_hidden_states:
lowercase : Optional[Any] = output + (hidden_states,)
return output
return BackboneOutput(feature_maps=snake_case ,hidden_states=snake_case ,attentions=snake_case )
| 20 |
"""simple docstring"""
import os
import re
import warnings
from shutil import copyfile
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
if TYPE_CHECKING:
from ...tokenization_utils_base import TextInput
from ...utils import logging
__SCREAMING_SNAKE_CASE : Union[str, Any] = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : List[Any] = {'vocab_file': 'spiece.model'}
__SCREAMING_SNAKE_CASE : int = {
'vocab_file': {
't5-small': 'https://huggingface.co/t5-small/resolve/main/spiece.model',
't5-base': 'https://huggingface.co/t5-base/resolve/main/spiece.model',
't5-large': 'https://huggingface.co/t5-large/resolve/main/spiece.model',
't5-3b': 'https://huggingface.co/t5-3b/resolve/main/spiece.model',
't5-11b': 'https://huggingface.co/t5-11b/resolve/main/spiece.model',
}
}
# TODO(PVP) - this should be removed in Transformers v5
__SCREAMING_SNAKE_CASE : Dict = {
't5-small': 512,
't5-base': 512,
't5-large': 512,
't5-3b': 512,
't5-11b': 512,
}
__SCREAMING_SNAKE_CASE : Optional[int] = '▁'
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = VOCAB_FILES_NAMES
__lowercase: Any = PRETRAINED_VOCAB_FILES_MAP
__lowercase: Any = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__lowercase: List[str] = ["""input_ids""", """attention_mask"""]
def __init__( self : Optional[int] , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Optional[Any]="</s>" , UpperCAmelCase_ : Optional[Any]="<unk>" , UpperCAmelCase_ : Any="<pad>" , UpperCAmelCase_ : Tuple=100 , UpperCAmelCase_ : Optional[Any]=None , UpperCAmelCase_ : Optional[Dict[str, Any]] = None , UpperCAmelCase_ : Optional[int]=True , **UpperCAmelCase_ : Dict , ) ->None:
"""simple docstring"""
if extra_ids > 0 and additional_special_tokens is None:
snake_case_ = [F"""<extra_id_{i}>""" for i in range(UpperCAmelCase_ )]
elif extra_ids > 0 and additional_special_tokens is not None:
# Check that we have the right number of extra_id special tokens
snake_case_ = len(set(filter(lambda UpperCAmelCase_ : bool("""extra_id""" in str(UpperCAmelCase_ ) ) , UpperCAmelCase_ ) ) )
if extra_tokens != extra_ids:
raise ValueError(
F"""Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are"""
""" provided to T5Tokenizer. In this case the additional_special_tokens must include the extra_ids"""
""" tokens""" )
if legacy:
logger.warning_once(
F"""You are using the legacy behaviour of the {self.__class__}. This means that tokens that come after special tokens will not be properly handled. We recommend you to"""
""" read the related pull request available at https://github.com/huggingface/transformers/pull/24565""" )
snake_case_ = legacy
snake_case_ = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
eos_token=UpperCAmelCase_ , unk_token=UpperCAmelCase_ , pad_token=UpperCAmelCase_ , extra_ids=UpperCAmelCase_ , additional_special_tokens=UpperCAmelCase_ , sp_model_kwargs=self.sp_model_kwargs , legacy=UpperCAmelCase_ , **UpperCAmelCase_ , )
snake_case_ = vocab_file
snake_case_ = extra_ids
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(UpperCAmelCase_ )
@staticmethod
def lowerCAmelCase ( UpperCAmelCase_ : str , UpperCAmelCase_ : Dict , UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
if pretrained_model_name_or_path in TaTokenizer.max_model_input_sizes:
snake_case_ = TaTokenizer.max_model_input_sizes[pretrained_model_name_or_path]
if init_max_model_length is not None and init_max_model_length != max_model_length:
return init_max_model_length
elif init_max_model_length is None:
warnings.warn(
"""This tokenizer was incorrectly instantiated with a model max length of"""
F""" {deprecated_max_model_length} which will be corrected in Transformers v5.\nFor now, this"""
""" behavior is kept to avoid breaking backwards compatibility when padding/encoding with"""
""" `truncation is True`.\n- Be aware that you SHOULD NOT rely on"""
F""" {pretrained_model_name_or_path} automatically truncating your input to"""
F""" {deprecated_max_model_length} when padding/encoding.\n- If you want to encode/pad to sequences"""
F""" longer than {deprecated_max_model_length} you can either instantiate this tokenizer with"""
""" `model_max_length` or pass `max_length` when encoding/padding.\n- To avoid this warning, please"""
""" instantiate this tokenizer with `model_max_length` set to your preferred value.""" , UpperCAmelCase_ , )
return max_model_length
@property
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
return self.sp_model.get_piece_size() + self._extra_ids
def lowerCAmelCase ( self : Any ) ->Optional[int]:
"""simple docstring"""
snake_case_ = {self.convert_ids_to_tokens(UpperCAmelCase_ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None , UpperCAmelCase_ : bool = False ) ->List[int]:
"""simple docstring"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCAmelCase_ , token_ids_a=UpperCAmelCase_ , already_has_special_tokens=UpperCAmelCase_ )
# normal case: some special tokens
if token_ids_a is None:
return ([0] * len(UpperCAmelCase_ )) + [1]
return ([0] * len(UpperCAmelCase_ )) + [1] + ([0] * len(UpperCAmelCase_ )) + [1]
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
return list(
set(filter(lambda UpperCAmelCase_ : bool(re.search(R"""<extra_id_\d+>""" , UpperCAmelCase_ ) ) is not None , self.additional_special_tokens ) ) )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
return [self._convert_token_to_id(UpperCAmelCase_ ) for token in self.get_sentinel_tokens()]
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : List[int] ) ->List[int]:
"""simple docstring"""
if len(UpperCAmelCase_ ) > 0 and token_ids[-1] == self.eos_token_id:
warnings.warn(
F"""This sequence already has {self.eos_token}. In future versions this behavior may lead to duplicated"""
""" eos tokens being added.""" )
return token_ids
else:
return token_ids + [self.eos_token_id]
def lowerCAmelCase ( self : str , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = [self.eos_token_id]
if token_ids_a is None:
return len(token_ids_a + eos ) * [0]
return len(token_ids_a + eos + token_ids_a + eos ) * [0]
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
if token_ids_a is None:
return token_ids_a
else:
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
return token_ids_a + token_ids_a
def __getstate__( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = self.__dict__.copy()
snake_case_ = None
return state
def __setstate__( self : Optional[Any] , UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = d
# for backward compatibility
if not hasattr(self , """sp_model_kwargs""" ):
snake_case_ = {}
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def lowerCAmelCase ( self : int , UpperCAmelCase_ : "TextInput" , **UpperCAmelCase_ : Tuple ) ->List[str]:
"""simple docstring"""
if not self.legacy:
snake_case_ = SPIECE_UNDERLINE + text.replace(UpperCAmelCase_ , """ """ )
return super().tokenize(UpperCAmelCase_ , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Any ) ->Tuple:
"""simple docstring"""
if not self.legacy:
snake_case_ = text.startswith(UpperCAmelCase_ )
if is_first:
snake_case_ = text[1:]
snake_case_ = self.sp_model.encode(UpperCAmelCase_ , out_type=UpperCAmelCase_ )
if not self.legacy and not is_first and not text.startswith(""" """ ) and tokens[0].startswith(UpperCAmelCase_ ):
snake_case_ = ([tokens[0][1:]] if len(tokens[0] ) > 1 else []) + tokens[1:]
return tokens
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
if token.startswith("""<extra_id_""" ):
snake_case_ = re.match(R"""<extra_id_(\d+)>""" , UpperCAmelCase_ )
snake_case_ = int(match.group(1 ) )
return self.vocab_size - num - 1
return self.sp_model.piece_to_id(UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : Optional[Any] ) ->List[Any]:
"""simple docstring"""
if index < self.sp_model.get_piece_size():
snake_case_ = self.sp_model.IdToPiece(UpperCAmelCase_ )
else:
snake_case_ = F"""<extra_id_{self.vocab_size - 1 - index}>"""
return token
def lowerCAmelCase ( self : List[Any] , UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = []
snake_case_ = """"""
snake_case_ = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(UpperCAmelCase_ ) + token
snake_case_ = True
snake_case_ = []
else:
current_sub_tokens.append(UpperCAmelCase_ )
snake_case_ = False
out_string += self.sp_model.decode(UpperCAmelCase_ )
return out_string.strip()
def lowerCAmelCase ( self : str , UpperCAmelCase_ : str , UpperCAmelCase_ : Optional[str] = None ) ->Tuple[str]:
"""simple docstring"""
if not os.path.isdir(UpperCAmelCase_ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case_ = os.path.join(
UpperCAmelCase_ , (filename_prefix + """-""" if filename_prefix else """""") + VOCAB_FILES_NAMES["""vocab_file"""] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(UpperCAmelCase_ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , UpperCAmelCase_ )
elif not os.path.isfile(self.vocab_file ):
with open(UpperCAmelCase_ , """wb""" ) as fi:
snake_case_ = self.sp_model.serialized_model_proto()
fi.write(UpperCAmelCase_ )
return (out_vocab_file,)
| 347 | 0 |
import datasets
import faiss
import numpy as np
import streamlit as st
import torch
from elasticsearch import Elasticsearch
from elia_utils import (
embed_questions_for_retrieval,
make_qa_sas_model,
qa_sas_generate,
query_es_index,
query_qa_dense_index,
)
import transformers
from transformers import AutoModel, AutoModelForSeqaSeqLM, AutoTokenizer
SCREAMING_SNAKE_CASE : str = "bart"
SCREAMING_SNAKE_CASE : Optional[int] = True
@st.cache(allow_output_mutation=lowerCamelCase_ )
def UpperCamelCase_( ) -> int:
if LOAD_DENSE_INDEX:
_lowercase : str = AutoTokenizer.from_pretrained('yjernite/retribert-base-uncased' )
_lowercase : Union[str, Any] = AutoModel.from_pretrained('yjernite/retribert-base-uncased' ).to('cuda:0' )
_lowercase : str = qar_model.eval()
else:
_lowercase , _lowercase : Any = (None, None)
if MODEL_TYPE == "bart":
_lowercase : Dict = AutoTokenizer.from_pretrained('yjernite/bart_eli5' )
_lowercase : int = AutoModelForSeqaSeqLM.from_pretrained('yjernite/bart_eli5' ).to('cuda:0' )
_lowercase : Any = torch.load('seq2seq_models/eli5_bart_model_blm_2.pth' )
sas_model.load_state_dict(save_dict['model'] )
_lowercase : List[Any] = sas_model.eval()
else:
_lowercase , _lowercase : Union[str, Any] = make_qa_sas_model(
model_name='t5-small' , from_file='seq2seq_models/eli5_t5_model_1024_4.pth' , device='cuda:0' )
return (qar_tokenizer, qar_model, sas_tokenizer, sas_model)
@st.cache(allow_output_mutation=lowerCamelCase_ )
def UpperCamelCase_( ) -> str:
if LOAD_DENSE_INDEX:
_lowercase : Optional[Any] = faiss.StandardGpuResources()
_lowercase : Optional[int] = datasets.load_dataset(path='wiki_snippets' , name='wiki40b_en_100_0' )['train']
_lowercase : Tuple = np.memmap(
'wiki40b_passages_reps_32_l-8_h-768_b-512-512.dat' , dtype='float32' , mode='r' , shape=(wikiaab_passages.num_rows, 128) , )
_lowercase : Any = faiss.IndexFlatIP(128 )
_lowercase : Union[str, Any] = faiss.index_cpu_to_gpu(lowerCamelCase_ , 1 , lowerCamelCase_ )
wikiaab_gpu_index_flat.add(lowerCamelCase_ ) # TODO fix for larger GPU
else:
_lowercase , _lowercase : Any = (None, None)
_lowercase : List[str] = Elasticsearch([{'host': 'localhost', 'port': '9200'}] )
return (wikiaab_passages, wikiaab_gpu_index_flat, es_client)
@st.cache(allow_output_mutation=lowerCamelCase_ )
def UpperCamelCase_( ) -> Any:
_lowercase : List[str] = datasets.load_dataset('eli5' , name='LFQA_reddit' )
_lowercase : Optional[Any] = elia['train_eli5']
_lowercase : Tuple = np.memmap(
'eli5_questions_reps.dat' , dtype='float32' , mode='r' , shape=(elia_train.num_rows, 128) )
_lowercase : Union[str, Any] = faiss.IndexFlatIP(128 )
eli5_train_q_index.add(lowerCamelCase_ )
return (elia_train, eli5_train_q_index)
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Any = load_indexes()
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Optional[Any] = load_models()
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : List[Any] = load_train_data()
def UpperCamelCase_( lowerCamelCase_ , lowerCamelCase_=10 ) -> List[str]:
_lowercase : Any = embed_questions_for_retrieval([question] , lowerCamelCase_ , lowerCamelCase_ )
_lowercase , _lowercase : List[str] = eli5_train_q_index.search(lowerCamelCase_ , lowerCamelCase_ )
_lowercase : List[str] = [elia_train[int(lowerCamelCase_ )] for i in I[0]]
return nn_examples
def UpperCamelCase_( lowerCamelCase_ , lowerCamelCase_="wiki40b" , lowerCamelCase_="dense" , lowerCamelCase_=10 ) -> Dict:
if source == "none":
_lowercase , _lowercase : Union[str, Any] = (' <P> '.join(['' for _ in range(11 )] ).strip(), [])
else:
if method == "dense":
_lowercase , _lowercase : Dict = query_qa_dense_index(
lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ )
else:
_lowercase , _lowercase : str = query_es_index(
lowerCamelCase_ , lowerCamelCase_ , index_name='english_wiki40b_snippets_100w' , n_results=lowerCamelCase_ , )
_lowercase : List[Any] = [
(res['article_title'], res['section_title'].strip(), res['score'], res['passage_text']) for res in hit_lst
]
_lowercase : Union[str, Any] = 'question: {} context: {}'.format(lowerCamelCase_ , lowerCamelCase_ )
return question_doc, support_list
@st.cache(
hash_funcs={
torch.Tensor: (lambda lowerCamelCase_ : None),
transformers.models.bart.tokenization_bart.BartTokenizer: (lambda lowerCamelCase_ : None),
} )
def UpperCamelCase_( lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_=64 , lowerCamelCase_=256 , lowerCamelCase_=False , lowerCamelCase_=2 , lowerCamelCase_=0.95 , lowerCamelCase_=0.8 ) -> Dict:
with torch.no_grad():
_lowercase : str = qa_sas_generate(
lowerCamelCase_ , lowerCamelCase_ , lowerCamelCase_ , num_answers=1 , num_beams=lowerCamelCase_ , min_len=lowerCamelCase_ , max_len=lowerCamelCase_ , do_sample=lowerCamelCase_ , temp=lowerCamelCase_ , top_p=lowerCamelCase_ , top_k=lowerCamelCase_ , max_input_length=1024 , device='cuda:0' , )[0]
return (answer, support_list)
st.title("Long Form Question Answering with ELI5")
# Start sidebar
SCREAMING_SNAKE_CASE : Union[str, Any] = "<img src='https://huggingface.co/front/assets/huggingface_logo.svg'>"
SCREAMING_SNAKE_CASE : List[Any] = "\n<html>\n <head>\n <style>\n .img-container {\n padding-left: 90px;\n padding-right: 90px;\n padding-top: 50px;\n padding-bottom: 50px;\n background-color: #f0f3f9;\n }\n </style>\n </head>\n <body>\n <span class=\"img-container\"> <!-- Inline parent element -->\n %s\n </span>\n </body>\n</html>\n" % (
header_html,
)
st.sidebar.markdown(
header_full,
unsafe_allow_html=True,
)
# Long Form QA with ELI5 and Wikipedia
SCREAMING_SNAKE_CASE : Any = "\nThis demo presents a model trained to [provide long-form answers to open-domain questions](https://yjernite.github.io/lfqa.html).\nFirst, a document retriever fetches a set of relevant Wikipedia passages given the question from the [Wiki40b](https://research.google/pubs/pub49029/) dataset,\na pre-processed fixed snapshot of Wikipedia.\n"
st.sidebar.markdown(description, unsafe_allow_html=True)
SCREAMING_SNAKE_CASE : Union[str, Any] = [
"Answer the question",
"View the retrieved document only",
"View the most similar ELI5 question and answer",
"Show me everything, please!",
]
SCREAMING_SNAKE_CASE : Optional[int] = st.sidebar.checkbox("Demo options")
if demo_options:
SCREAMING_SNAKE_CASE : List[str] = st.sidebar.selectbox(
"",
action_list,
index=3,
)
SCREAMING_SNAKE_CASE : Optional[int] = action_list.index(action_st)
SCREAMING_SNAKE_CASE : Tuple = st.sidebar.selectbox(
"",
["Show full text of passages", "Show passage section titles"],
index=0,
)
SCREAMING_SNAKE_CASE : int = show_type == "Show full text of passages"
else:
SCREAMING_SNAKE_CASE : Any = 3
SCREAMING_SNAKE_CASE : Dict = True
SCREAMING_SNAKE_CASE : Union[str, Any] = st.sidebar.checkbox("Retrieval options")
if retrieval_options:
SCREAMING_SNAKE_CASE : Tuple = "\n ### Information retriever options\n\n The **sparse** retriever uses ElasticSearch, while the **dense** retriever uses max-inner-product search between a question and passage embedding\n trained using the [ELI5](https://arxiv.org/abs/1907.09190) questions-answer pairs.\n The answer is then generated by sequence to sequence model which takes the question and retrieved document as input.\n "
st.sidebar.markdown(retriever_info)
SCREAMING_SNAKE_CASE : Dict = st.sidebar.selectbox("Which Wikipedia format should the model use?", ["wiki40b", "none"])
SCREAMING_SNAKE_CASE : Union[str, Any] = st.sidebar.selectbox("Which Wikipedia indexer should the model use?", ["dense", "sparse", "mixed"])
else:
SCREAMING_SNAKE_CASE : int = "wiki40b"
SCREAMING_SNAKE_CASE : int = "dense"
SCREAMING_SNAKE_CASE : str = "beam"
SCREAMING_SNAKE_CASE : Optional[Any] = 2
SCREAMING_SNAKE_CASE : List[str] = 64
SCREAMING_SNAKE_CASE : Union[str, Any] = 256
SCREAMING_SNAKE_CASE : Union[str, Any] = None
SCREAMING_SNAKE_CASE : List[Any] = None
SCREAMING_SNAKE_CASE : str = st.sidebar.checkbox("Generation options")
if generate_options:
SCREAMING_SNAKE_CASE : Any = "\n ### Answer generation options\n\n The sequence-to-sequence model was initialized with [BART](https://huggingface.co/facebook/bart-large)\n weights and fine-tuned on the ELI5 QA pairs and retrieved documents. You can use the model for greedy decoding with\n **beam** search, or **sample** from the decoder's output probabilities.\n "
st.sidebar.markdown(generate_info)
SCREAMING_SNAKE_CASE : List[Any] = st.sidebar.selectbox("Would you like to use beam search or sample an answer?", ["beam", "sampled"])
SCREAMING_SNAKE_CASE : Tuple = st.sidebar.slider(
"Minimum generation length", min_value=8, max_value=256, value=64, step=8, format=None, key=None
)
SCREAMING_SNAKE_CASE : int = st.sidebar.slider(
"Maximum generation length", min_value=64, max_value=512, value=256, step=16, format=None, key=None
)
if sampled == "beam":
SCREAMING_SNAKE_CASE : int = st.sidebar.slider("Beam size", min_value=1, max_value=8, value=2, step=None, format=None, key=None)
else:
SCREAMING_SNAKE_CASE : Union[str, Any] = st.sidebar.slider(
"Nucleus sampling p", min_value=0.1, max_value=1.0, value=0.95, step=0.01, format=None, key=None
)
SCREAMING_SNAKE_CASE : Any = st.sidebar.slider(
"Temperature", min_value=0.1, max_value=1.0, value=0.7, step=0.01, format=None, key=None
)
SCREAMING_SNAKE_CASE : str = None
# start main text
SCREAMING_SNAKE_CASE : List[str] = [
"<MY QUESTION>",
"How do people make chocolate?",
"Why do we get a fever when we are sick?",
"How can different animals perceive different colors?",
"What is natural language processing?",
"What's the best way to treat a sunburn?",
"What exactly are vitamins ?",
"How does nuclear energy provide electricity?",
"What's the difference between viruses and bacteria?",
"Why are flutes classified as woodwinds when most of them are made out of metal ?",
"Why do people like drinking coffee even though it tastes so bad?",
"What happens when wine ages? How does it make the wine taste better?",
"If an animal is an herbivore, where does it get the protein that it needs to survive if it only eats grass?",
"How can we set a date to the beginning or end of an artistic period? Doesn't the change happen gradually?",
"How does New Zealand have so many large bird predators?",
]
SCREAMING_SNAKE_CASE : str = st.selectbox(
"What would you like to ask? ---- select <MY QUESTION> to enter a new query",
questions_list,
index=1,
)
if question_s == "<MY QUESTION>":
SCREAMING_SNAKE_CASE : List[str] = st.text_input("Enter your question here:", "")
else:
SCREAMING_SNAKE_CASE : Optional[int] = question_s
if st.button("Show me!"):
if action in [0, 1, 3]:
if index_type == "mixed":
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Optional[int] = make_support(question, source=wiki_source, method="dense", n_results=10)
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Any = make_support(question, source=wiki_source, method="sparse", n_results=10)
SCREAMING_SNAKE_CASE : Tuple = []
for res_d, res_s in zip(support_list_dense, support_list_sparse):
if tuple(res_d) not in support_list:
support_list += [tuple(res_d)]
if tuple(res_s) not in support_list:
support_list += [tuple(res_s)]
SCREAMING_SNAKE_CASE : Optional[Any] = support_list[:10]
SCREAMING_SNAKE_CASE : int = "<P> " + " <P> ".join([res[-1] for res in support_list])
else:
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Any = make_support(question, source=wiki_source, method=index_type, n_results=10)
if action in [0, 3]:
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE : Tuple = answer_question(
question_doc,
sas_model,
sas_tokenizer,
min_len=min_len,
max_len=int(max_len),
sampling=(sampled == "sampled"),
n_beams=n_beams,
top_p=top_p,
temp=temp,
)
st.markdown("### The model generated answer is:")
st.write(answer)
if action in [0, 1, 3] and wiki_source != "none":
st.markdown("--- \n ### The model is drawing information from the following Wikipedia passages:")
for i, res in enumerate(support_list):
SCREAMING_SNAKE_CASE : Optional[Any] = "https://en.wikipedia.org/wiki/{}".format(res[0].replace(" ", "_"))
SCREAMING_SNAKE_CASE : List[Any] = res[1].strip()
if sec_titles == "":
SCREAMING_SNAKE_CASE : Union[str, Any] = "[{}]({})".format(res[0], wiki_url)
else:
SCREAMING_SNAKE_CASE : Any = sec_titles.split(" & ")
SCREAMING_SNAKE_CASE : List[Any] = " & ".join(
["[{}]({}#{})".format(sec.strip(), wiki_url, sec.strip().replace(" ", "_")) for sec in sec_list]
)
st.markdown(
"{0:02d} - **Article**: {1:<18} <br> _Section_: {2}".format(i + 1, res[0], sections),
unsafe_allow_html=True,
)
if show_passages:
st.write(
"> <span style=\"font-family:arial; font-size:10pt;\">" + res[-1] + "</span>", unsafe_allow_html=True
)
if action in [2, 3]:
SCREAMING_SNAKE_CASE : str = find_nearest_training(question)
SCREAMING_SNAKE_CASE : Any = nn_train_list[0]
st.markdown(
"--- \n ### The most similar question in the ELI5 training set was: \n\n {}".format(train_exple["title"])
)
SCREAMING_SNAKE_CASE : str = [
"{}. {}".format(i + 1, " \n".join([line.strip() for line in ans.split("\n") if line.strip() != ""]))
for i, (ans, sc) in enumerate(zip(train_exple["answers"]["text"], train_exple["answers"]["score"]))
if i == 0 or sc > 2
]
st.markdown("##### Its answers were: \n\n {}".format("\n".join(answers_st)))
SCREAMING_SNAKE_CASE : Tuple = "\n---\n\n**Disclaimer**\n\n*The intent of this app is to provide some (hopefully entertaining) insights into the behavior of a current LFQA system.\nEvaluating biases of such a model and ensuring factual generations are still very much open research problems.\nTherefore, until some significant progress is achieved, we caution against using the generated answers for practical purposes.*\n"
st.sidebar.markdown(disclaimer, unsafe_allow_html=True)
| 21 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE = 1_000_000 ) -> int:
snake_case_ = [i - 1 for i in range(limit + 1 )]
for i in range(2 , limit + 1 ):
if phi[i] == i - 1:
for j in range(2 * i , limit + 1 , _SCREAMING_SNAKE_CASE ):
phi[j] -= phi[j] // i
return sum(phi[2 : limit + 1] )
if __name__ == "__main__":
print(solution())
| 347 | 0 |
'''simple docstring'''
def UpperCAmelCase_ ( __lowercase : str , __lowercase : str ) -> bool:
'''simple docstring'''
_UpperCAmelCase = len(__lowercase ) + 1
_UpperCAmelCase = len(__lowercase ) + 1
# dp is a 2d matrix where dp[i][j] denotes whether prefix string of
# length i of input_string matches with prefix string of length j of
# given pattern.
# "dp" stands for dynamic programming.
_UpperCAmelCase = [[0 for i in range(__lowercase )] for j in range(__lowercase )]
# since string of zero length match pattern of zero length
_UpperCAmelCase = 1
# since pattern of zero length will never match with string of non-zero length
for i in range(1 , __lowercase ):
_UpperCAmelCase = 0
# since string of zero length will match with pattern where there
# is at least one * alternatively
for j in range(1 , __lowercase ):
_UpperCAmelCase = dp[0][j - 2] if pattern[j - 1] == "*" else 0
# now using bottom-up approach to find for all remaining lengths
for i in range(1 , __lowercase ):
for j in range(1 , __lowercase ):
if input_string[i - 1] == pattern[j - 1] or pattern[j - 1] == ".":
_UpperCAmelCase = dp[i - 1][j - 1]
elif pattern[j - 1] == "*":
if dp[i][j - 2] == 1:
_UpperCAmelCase = 1
elif pattern[j - 2] in (input_string[i - 1], "."):
_UpperCAmelCase = dp[i - 1][j]
else:
_UpperCAmelCase = 0
else:
_UpperCAmelCase = 0
return bool(dp[-1][-1] )
if __name__ == "__main__":
import doctest
doctest.testmod()
# inputing the strings
# input_string = input("input a string :")
# pattern = input("input a pattern :")
__SCREAMING_SNAKE_CASE :str = '''aab'''
__SCREAMING_SNAKE_CASE :Optional[Any] = '''c*a*b'''
# using function to check whether given string matches the given pattern
if match_pattern(input_string, pattern):
print(F"{input_string} matches the given pattern {pattern}")
else:
print(F"{input_string} does not match with the given pattern {pattern}")
| 22 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
print(f"""Vertex\tShortest Distance from vertex {src}""" )
for i, d in enumerate(_SCREAMING_SNAKE_CASE ):
print(f"""{i}\t\t{d}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
return True
return False
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> list[float]:
snake_case_ = [float("""inf""" )] * vertex_count
snake_case_ = 0.0
for _ in range(vertex_count - 1 ):
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
snake_case_ = distance[u] + w
snake_case_ = check_negative_cycle(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if negative_cycle_exists:
raise Exception("""Negative cycle found""" )
return distance
if __name__ == "__main__":
import doctest
doctest.testmod()
__SCREAMING_SNAKE_CASE : int = int(input('Enter number of vertices: ').strip())
__SCREAMING_SNAKE_CASE : Dict = int(input('Enter number of edges: ').strip())
__SCREAMING_SNAKE_CASE : list[dict[str, int]] = [{} for _ in range(E)]
for i in range(E):
print('Edge ', i + 1)
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE : List[str] = (
int(x)
for x in input('Enter source, destination, weight: ').strip().split(' ')
)
__SCREAMING_SNAKE_CASE : Union[str, Any] = {'src': src, 'dst': dest, 'weight': weight}
__SCREAMING_SNAKE_CASE : Union[str, Any] = int(input('\nEnter shortest path source:').strip())
__SCREAMING_SNAKE_CASE : str = bellman_ford(graph, V, E, source)
print_distance(shortest_distance, 0)
| 347 | 0 |
'''simple docstring'''
from math import cos, sin, sqrt, tau
from audio_filters.iir_filter import IIRFilter
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float = 1 / sqrt(2 ) ) -> IIRFilter:
UpperCAmelCase : Optional[int] = tau * frequency / samplerate
UpperCAmelCase : List[Any] = sin(_lowerCAmelCase )
UpperCAmelCase : Optional[Any] = cos(_lowerCAmelCase )
UpperCAmelCase : int = _sin / (2 * q_factor)
UpperCAmelCase : Any = (1 - _cos) / 2
UpperCAmelCase : List[Any] = 1 - _cos
UpperCAmelCase : Union[str, Any] = 1 + alpha
UpperCAmelCase : Any = -2 * _cos
UpperCAmelCase : Dict = 1 - alpha
UpperCAmelCase : Union[str, Any] = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float = 1 / sqrt(2 ) ) -> IIRFilter:
UpperCAmelCase : Any = tau * frequency / samplerate
UpperCAmelCase : Tuple = sin(_lowerCAmelCase )
UpperCAmelCase : Tuple = cos(_lowerCAmelCase )
UpperCAmelCase : Dict = _sin / (2 * q_factor)
UpperCAmelCase : int = (1 + _cos) / 2
UpperCAmelCase : List[Any] = -1 - _cos
UpperCAmelCase : Tuple = 1 + alpha
UpperCAmelCase : List[str] = -2 * _cos
UpperCAmelCase : Optional[Any] = 1 - alpha
UpperCAmelCase : Tuple = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float = 1 / sqrt(2 ) ) -> IIRFilter:
UpperCAmelCase : Optional[int] = tau * frequency / samplerate
UpperCAmelCase : Optional[int] = sin(_lowerCAmelCase )
UpperCAmelCase : Tuple = cos(_lowerCAmelCase )
UpperCAmelCase : Optional[int] = _sin / (2 * q_factor)
UpperCAmelCase : Union[str, Any] = _sin / 2
UpperCAmelCase : Any = 0
UpperCAmelCase : int = -ba
UpperCAmelCase : Optional[Any] = 1 + alpha
UpperCAmelCase : List[Any] = -2 * _cos
UpperCAmelCase : Optional[Any] = 1 - alpha
UpperCAmelCase : int = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float = 1 / sqrt(2 ) ) -> IIRFilter:
UpperCAmelCase : List[str] = tau * frequency / samplerate
UpperCAmelCase : Union[str, Any] = sin(_lowerCAmelCase )
UpperCAmelCase : str = cos(_lowerCAmelCase )
UpperCAmelCase : Optional[Any] = _sin / (2 * q_factor)
UpperCAmelCase : List[str] = 1 - alpha
UpperCAmelCase : Any = -2 * _cos
UpperCAmelCase : Optional[int] = 1 + alpha
UpperCAmelCase : Union[str, Any] = IIRFilter(2 )
filt.set_coefficients([ba, ba, ba] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float , _lowerCAmelCase : float = 1 / sqrt(2 ) , ) -> IIRFilter:
UpperCAmelCase : Optional[Any] = tau * frequency / samplerate
UpperCAmelCase : Union[str, Any] = sin(_lowerCAmelCase )
UpperCAmelCase : Optional[int] = cos(_lowerCAmelCase )
UpperCAmelCase : Dict = _sin / (2 * q_factor)
UpperCAmelCase : str = 10 ** (gain_db / 40)
UpperCAmelCase : int = 1 + alpha * big_a
UpperCAmelCase : Union[str, Any] = -2 * _cos
UpperCAmelCase : Optional[Any] = 1 - alpha * big_a
UpperCAmelCase : Union[str, Any] = 1 + alpha / big_a
UpperCAmelCase : Tuple = -2 * _cos
UpperCAmelCase : Any = 1 - alpha / big_a
UpperCAmelCase : Optional[Any] = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float , _lowerCAmelCase : float = 1 / sqrt(2 ) , ) -> IIRFilter:
UpperCAmelCase : Any = tau * frequency / samplerate
UpperCAmelCase : Optional[int] = sin(_lowerCAmelCase )
UpperCAmelCase : Union[str, Any] = cos(_lowerCAmelCase )
UpperCAmelCase : str = _sin / (2 * q_factor)
UpperCAmelCase : List[str] = 10 ** (gain_db / 40)
UpperCAmelCase : Optional[int] = (big_a + 1) - (big_a - 1) * _cos
UpperCAmelCase : int = (big_a + 1) + (big_a - 1) * _cos
UpperCAmelCase : int = (big_a - 1) - (big_a + 1) * _cos
UpperCAmelCase : Optional[int] = (big_a - 1) + (big_a + 1) * _cos
UpperCAmelCase : str = 2 * sqrt(_lowerCAmelCase ) * alpha
UpperCAmelCase : Dict = big_a * (pmc + aaa)
UpperCAmelCase : Any = 2 * big_a * mpc
UpperCAmelCase : Union[str, Any] = big_a * (pmc - aaa)
UpperCAmelCase : Optional[int] = ppmc + aaa
UpperCAmelCase : Optional[Any] = -2 * pmpc
UpperCAmelCase : Optional[Any] = ppmc - aaa
UpperCAmelCase : int = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
def snake_case_ ( _lowerCAmelCase : int , _lowerCAmelCase : int , _lowerCAmelCase : float , _lowerCAmelCase : float = 1 / sqrt(2 ) , ) -> IIRFilter:
UpperCAmelCase : int = tau * frequency / samplerate
UpperCAmelCase : Union[str, Any] = sin(_lowerCAmelCase )
UpperCAmelCase : Union[str, Any] = cos(_lowerCAmelCase )
UpperCAmelCase : Any = _sin / (2 * q_factor)
UpperCAmelCase : int = 10 ** (gain_db / 40)
UpperCAmelCase : List[str] = (big_a + 1) - (big_a - 1) * _cos
UpperCAmelCase : Union[str, Any] = (big_a + 1) + (big_a - 1) * _cos
UpperCAmelCase : Optional[Any] = (big_a - 1) - (big_a + 1) * _cos
UpperCAmelCase : Union[str, Any] = (big_a - 1) + (big_a + 1) * _cos
UpperCAmelCase : List[str] = 2 * sqrt(_lowerCAmelCase ) * alpha
UpperCAmelCase : Any = big_a * (ppmc + aaa)
UpperCAmelCase : str = -2 * big_a * pmpc
UpperCAmelCase : List[Any] = big_a * (ppmc - aaa)
UpperCAmelCase : Optional[Any] = pmc + aaa
UpperCAmelCase : Any = 2 * mpc
UpperCAmelCase : str = pmc - aaa
UpperCAmelCase : Union[str, Any] = IIRFilter(2 )
filt.set_coefficients([aa, aa, aa] , [ba, ba, ba] )
return filt
| 23 |
"""simple docstring"""
import argparse
import logging
import os
import re
import tensorflow as tf
from transformers import (
AutoConfig,
AutoTokenizer,
DataCollatorForLanguageModeling,
PushToHubCallback,
TFAutoModelForMaskedLM,
create_optimizer,
)
__SCREAMING_SNAKE_CASE : List[str] = logging.getLogger(__name__)
__SCREAMING_SNAKE_CASE : str = tf.data.AUTOTUNE
def _a ( ) -> List[str]:
snake_case_ = argparse.ArgumentParser(description="""Train a masked language model on TPU.""" )
parser.add_argument(
"""--pretrained_model_config""" , type=_SCREAMING_SNAKE_CASE , default="""roberta-base""" , help="""The model config to use. Note that we don't copy the model's weights, only the config!""" , )
parser.add_argument(
"""--tokenizer""" , type=_SCREAMING_SNAKE_CASE , default="""unigram-tokenizer-wikitext""" , help="""The name of the tokenizer to load. We use the pretrained tokenizer to initialize the model's vocab size.""" , )
parser.add_argument(
"""--per_replica_batch_size""" , type=_SCREAMING_SNAKE_CASE , default=8 , help="""Batch size per TPU core.""" , )
parser.add_argument(
"""--no_tpu""" , action="""store_true""" , help="""If set, run on CPU and don't try to initialize a TPU. Useful for debugging on non-TPU instances.""" , )
parser.add_argument(
"""--tpu_name""" , type=_SCREAMING_SNAKE_CASE , help="""Name of TPU resource to initialize. Should be blank on Colab, and 'local' on TPU VMs.""" , default="""local""" , )
parser.add_argument(
"""--tpu_zone""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud zone that TPU resource is located in. Only used for non-Colab TPU nodes.""" , )
parser.add_argument(
"""--gcp_project""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud project name. Only used for non-Colab TPU nodes.""" )
parser.add_argument(
"""--bfloat16""" , action="""store_true""" , help="""Use mixed-precision bfloat16 for training. This is the recommended lower-precision format for TPU.""" , )
parser.add_argument(
"""--train_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to training dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--shuffle_buffer_size""" , type=_SCREAMING_SNAKE_CASE , default=2**18 , help="""Size of the shuffle buffer (in samples)""" , )
parser.add_argument(
"""--eval_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to evaluation dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--num_epochs""" , type=_SCREAMING_SNAKE_CASE , default=1 , help="""Number of epochs to train for.""" , )
parser.add_argument(
"""--learning_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-4 , help="""Learning rate to use for training.""" , )
parser.add_argument(
"""--weight_decay_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-3 , help="""Weight decay rate to use for training.""" , )
parser.add_argument(
"""--max_length""" , type=_SCREAMING_SNAKE_CASE , default=512 , help="""Maximum length of tokenized sequences. Should match the setting used in prepare_tfrecord_shards.py""" , )
parser.add_argument(
"""--mlm_probability""" , type=_SCREAMING_SNAKE_CASE , default=0.15 , help="""Fraction of tokens to mask during training.""" , )
parser.add_argument("""--output_dir""" , type=_SCREAMING_SNAKE_CASE , required=_SCREAMING_SNAKE_CASE , help="""Path to save model checkpoints to.""" )
parser.add_argument("""--hub_model_id""" , type=_SCREAMING_SNAKE_CASE , help="""Model ID to upload to on the Hugging Face Hub.""" )
snake_case_ = parser.parse_args()
return args
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
try:
if args.tpu_name:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver(
args.tpu_name , zone=args.tpu_zone , project=args.gcp_project )
else:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
raise RuntimeError(
"""Couldn't connect to TPU! Most likely you need to specify --tpu_name, --tpu_zone, or """
"""--gcp_project. When running on a TPU VM, use --tpu_name local.""" )
tf.config.experimental_connect_to_cluster(_SCREAMING_SNAKE_CASE )
tf.tpu.experimental.initialize_tpu_system(_SCREAMING_SNAKE_CASE )
return tpu
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = 0
for file in file_list:
snake_case_ = file.split("""/""" )[-1]
snake_case_ = re.search(r"""-\d+-(\d+)\.tfrecord""" , _SCREAMING_SNAKE_CASE ).group(1 )
snake_case_ = int(_SCREAMING_SNAKE_CASE )
num_samples += sample_count
return num_samples
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None ) -> Union[str, Any]:
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.data.Dataset.from_tensor_slices(_SCREAMING_SNAKE_CASE )
if shuffle:
snake_case_ = dataset.shuffle(len(_SCREAMING_SNAKE_CASE ) )
snake_case_ = tf.data.TFRecordDataset(_SCREAMING_SNAKE_CASE , num_parallel_reads=_SCREAMING_SNAKE_CASE )
# TF can't infer the total sample count because it doesn't read all the records yet, so we assert it here
snake_case_ = dataset.apply(tf.data.experimental.assert_cardinality(_SCREAMING_SNAKE_CASE ) )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
if shuffle:
assert shuffle_buffer_size is not None
snake_case_ = dataset.shuffle(args.shuffle_buffer_size )
snake_case_ = dataset.batch(_SCREAMING_SNAKE_CASE , drop_remainder=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.prefetch(_SCREAMING_SNAKE_CASE )
return dataset
def _a ( _SCREAMING_SNAKE_CASE ) -> List[Any]:
if not args.no_tpu:
snake_case_ = initialize_tpu(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.distribute.TPUStrategy(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = tf.distribute.OneDeviceStrategy(device="""/gpu:0""" )
if args.bfloataa:
tf.keras.mixed_precision.set_global_policy("""mixed_bfloat16""" )
snake_case_ = AutoTokenizer.from_pretrained(args.tokenizer )
snake_case_ = AutoConfig.from_pretrained(args.pretrained_model_config )
snake_case_ = tokenizer.vocab_size
snake_case_ = tf.io.gfile.glob(os.path.join(args.train_dataset , """*.tfrecord""" ) )
if not training_records:
raise ValueError(f"""No .tfrecord files found in {args.train_dataset}.""" )
snake_case_ = tf.io.gfile.glob(os.path.join(args.eval_dataset , """*.tfrecord""" ) )
if not eval_records:
raise ValueError(f"""No .tfrecord files found in {args.eval_dataset}.""" )
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = num_train_samples // (args.per_replica_batch_size * strategy.num_replicas_in_sync)
snake_case_ = steps_per_epoch * args.num_epochs
with strategy.scope():
snake_case_ = TFAutoModelForMaskedLM.from_config(_SCREAMING_SNAKE_CASE )
model(model.dummy_inputs ) # Pass some dummy inputs through the model to ensure all the weights are built
snake_case_ , snake_case_ = create_optimizer(
num_train_steps=_SCREAMING_SNAKE_CASE , num_warmup_steps=total_train_steps // 20 , init_lr=args.learning_rate , weight_decay_rate=args.weight_decay_rate , )
# Transformers models compute the right loss for their task by default when labels are passed, and will
# use this for training unless you specify your own loss function in compile().
model.compile(optimizer=_SCREAMING_SNAKE_CASE , metrics=["""accuracy"""] )
def decode_fn(_SCREAMING_SNAKE_CASE ):
snake_case_ = {
"""input_ids""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
"""attention_mask""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
}
return tf.io.parse_single_example(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# Many of the data collators in Transformers are TF-compilable when return_tensors == "tf", so we can
# use their methods in our data pipeline.
snake_case_ = DataCollatorForLanguageModeling(
tokenizer=_SCREAMING_SNAKE_CASE , mlm_probability=args.mlm_probability , mlm=_SCREAMING_SNAKE_CASE , return_tensors="""tf""" )
def mask_with_collator(_SCREAMING_SNAKE_CASE ):
# TF really needs an isin() function
snake_case_ = (
~tf.cast(batch["""attention_mask"""] , tf.bool )
| (batch["""input_ids"""] == tokenizer.cls_token_id)
| (batch["""input_ids"""] == tokenizer.sep_token_id)
)
snake_case_ , snake_case_ = data_collator.tf_mask_tokens(
batch["""input_ids"""] , vocab_size=len(_SCREAMING_SNAKE_CASE ) , mask_token_id=tokenizer.mask_token_id , special_tokens_mask=_SCREAMING_SNAKE_CASE , )
return batch
snake_case_ = args.per_replica_batch_size * strategy.num_replicas_in_sync
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , shuffle_buffer_size=args.shuffle_buffer_size , )
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , )
snake_case_ = []
if args.hub_model_id:
callbacks.append(
PushToHubCallback(output_dir=args.output_dir , hub_model_id=args.hub_model_id , tokenizer=_SCREAMING_SNAKE_CASE ) )
model.fit(
_SCREAMING_SNAKE_CASE , validation_data=_SCREAMING_SNAKE_CASE , epochs=args.num_epochs , callbacks=_SCREAMING_SNAKE_CASE , )
model.save_pretrained(args.output_dir )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : Union[str, Any] = parse_args()
main(args)
| 347 | 0 |
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_torch_available,
is_vision_available,
)
snake_case_ = {'configuration_beit': ['BEIT_PRETRAINED_CONFIG_ARCHIVE_MAP', 'BeitConfig', 'BeitOnnxConfig']}
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ = ['BeitFeatureExtractor']
snake_case_ = ['BeitImageProcessor']
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ = [
'BEIT_PRETRAINED_MODEL_ARCHIVE_LIST',
'BeitForImageClassification',
'BeitForMaskedImageModeling',
'BeitForSemanticSegmentation',
'BeitModel',
'BeitPreTrainedModel',
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case_ = [
'FlaxBeitForImageClassification',
'FlaxBeitForMaskedImageModeling',
'FlaxBeitModel',
'FlaxBeitPreTrainedModel',
]
if TYPE_CHECKING:
from .configuration_beit import BEIT_PRETRAINED_CONFIG_ARCHIVE_MAP, BeitConfig, BeitOnnxConfig
try:
if not is_vision_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .feature_extraction_beit import BeitFeatureExtractor
from .image_processing_beit import BeitImageProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_beit import (
BEIT_PRETRAINED_MODEL_ARCHIVE_LIST,
BeitForImageClassification,
BeitForMaskedImageModeling,
BeitForSemanticSegmentation,
BeitModel,
BeitPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_beit import (
FlaxBeitForImageClassification,
FlaxBeitForMaskedImageModeling,
FlaxBeitModel,
FlaxBeitPreTrainedModel,
)
else:
import sys
snake_case_ = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 24 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> float:
if density <= 0:
raise ValueError("""Impossible fluid density""" )
if bulk_modulus <= 0:
raise ValueError("""Impossible bulk modulus""" )
return (bulk_modulus / density) ** 0.5
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_sentencepiece_available,
is_tf_available,
is_tokenizers_available,
is_torch_available,
)
UpperCAmelCase__ : str = {
'configuration_xlm_roberta': [
'XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP',
'XLMRobertaConfig',
'XLMRobertaOnnxConfig',
],
}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
UpperCAmelCase__ : Optional[int] = ['XLMRobertaTokenizer']
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
UpperCAmelCase__ : Optional[Any] = ['XLMRobertaTokenizerFast']
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
UpperCAmelCase__ : Tuple = [
'XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST',
'XLMRobertaForCausalLM',
'XLMRobertaForMaskedLM',
'XLMRobertaForMultipleChoice',
'XLMRobertaForQuestionAnswering',
'XLMRobertaForSequenceClassification',
'XLMRobertaForTokenClassification',
'XLMRobertaModel',
'XLMRobertaPreTrainedModel',
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
UpperCAmelCase__ : Tuple = [
'TF_XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST',
'TFXLMRobertaForCausalLM',
'TFXLMRobertaForMaskedLM',
'TFXLMRobertaForMultipleChoice',
'TFXLMRobertaForQuestionAnswering',
'TFXLMRobertaForSequenceClassification',
'TFXLMRobertaForTokenClassification',
'TFXLMRobertaModel',
'TFXLMRobertaPreTrainedModel',
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
UpperCAmelCase__ : Optional[int] = [
'FLAX_XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST',
'FlaxXLMRobertaForMaskedLM',
'FlaxXLMRobertaForCausalLM',
'FlaxXLMRobertaForMultipleChoice',
'FlaxXLMRobertaForQuestionAnswering',
'FlaxXLMRobertaForSequenceClassification',
'FlaxXLMRobertaForTokenClassification',
'FlaxXLMRobertaModel',
'FlaxXLMRobertaPreTrainedModel',
]
if TYPE_CHECKING:
from .configuration_xlm_roberta import (
XLM_ROBERTA_PRETRAINED_CONFIG_ARCHIVE_MAP,
XLMRobertaConfig,
XLMRobertaOnnxConfig,
)
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_xlm_roberta import XLMRobertaTokenizer
try:
if not is_tokenizers_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_xlm_roberta_fast import XLMRobertaTokenizerFast
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_xlm_roberta import (
XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
XLMRobertaForCausalLM,
XLMRobertaForMaskedLM,
XLMRobertaForMultipleChoice,
XLMRobertaForQuestionAnswering,
XLMRobertaForSequenceClassification,
XLMRobertaForTokenClassification,
XLMRobertaModel,
XLMRobertaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_xlm_roberta import (
TF_XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
TFXLMRobertaForCausalLM,
TFXLMRobertaForMaskedLM,
TFXLMRobertaForMultipleChoice,
TFXLMRobertaForQuestionAnswering,
TFXLMRobertaForSequenceClassification,
TFXLMRobertaForTokenClassification,
TFXLMRobertaModel,
TFXLMRobertaPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_flax_xlm_roberta import (
FLAX_XLM_ROBERTA_PRETRAINED_MODEL_ARCHIVE_LIST,
FlaxXLMRobertaForCausalLM,
FlaxXLMRobertaForMaskedLM,
FlaxXLMRobertaForMultipleChoice,
FlaxXLMRobertaForQuestionAnswering,
FlaxXLMRobertaForSequenceClassification,
FlaxXLMRobertaForTokenClassification,
FlaxXLMRobertaModel,
FlaxXLMRobertaPreTrainedModel,
)
else:
import sys
UpperCAmelCase__ : List[str] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 25 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE ) -> bool:
if num < 0:
return False
snake_case_ = num
snake_case_ = 0
while num > 0:
snake_case_ = rev_num * 10 + (num % 10)
num //= 10
return num_copy == rev_num
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
# Lint as: python3
# pylint: enable=line-too-long
# pylint: disable=g-import-not-at-top,g-bad-import-order,wrong-import-position
_snake_case = "2.13.1"
import platform
import pyarrow
from packaging import version
if version.parse(platform.python_version()) < version.parse("3.7"):
raise ImportWarning(
"To use `datasets`, Python>=3.7 is required, and the current version of Python doesn't match this condition."
)
if version.parse(pyarrow.__version__).major < 8:
raise ImportWarning(
"To use `datasets`, the module `pyarrow>=8.0.0` is required, and the current version of `pyarrow` doesn't match this condition.\n"
"If you are running this in a Google Colab, you should probably just restart the runtime to use the right version of `pyarrow`."
)
del platform
del pyarrow
del version
from .arrow_dataset import Dataset
from .arrow_reader import ReadInstruction
from .builder import ArrowBasedBuilder, BeamBasedBuilder, BuilderConfig, DatasetBuilder, GeneratorBasedBuilder
from .combine import concatenate_datasets, interleave_datasets
from .dataset_dict import DatasetDict, IterableDatasetDict
from .download import *
from .features import *
from .fingerprint import disable_caching, enable_caching, is_caching_enabled, set_caching_enabled
from .info import DatasetInfo, MetricInfo
from .inspect import (
get_dataset_config_info,
get_dataset_config_names,
get_dataset_infos,
get_dataset_split_names,
inspect_dataset,
inspect_metric,
list_datasets,
list_metrics,
)
from .iterable_dataset import IterableDataset
from .load import load_dataset, load_dataset_builder, load_from_disk, load_metric
from .metric import Metric
from .splits import (
NamedSplit,
NamedSplitAll,
Split,
SplitBase,
SplitDict,
SplitGenerator,
SplitInfo,
SubSplitInfo,
percent,
)
from .tasks import *
from .utils import *
from .utils import logging
# deprecated modules
from datasets import arrow_dataset as _arrow_dataset # isort:skip
from datasets import utils as _utils # isort:skip
from datasets.utils import download_manager as _deprecated_download_manager # isort:skip
_snake_case = concatenate_datasets
_snake_case = DownloadConfig
_snake_case = DownloadManager
_snake_case = DownloadMode
_snake_case = DownloadConfig
_snake_case = DownloadMode
_snake_case = DownloadManager
del _arrow_dataset, _utils, _deprecated_download_manager
| 26 |
"""simple docstring"""
import unittest
from transformers import SPIECE_UNDERLINE
from transformers.models.speechta import SpeechTaTokenizer
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.tokenization_utils import AddedToken
from ...test_tokenization_common import TokenizerTesterMixin
__SCREAMING_SNAKE_CASE : Tuple = get_tests_dir('fixtures/test_sentencepiece_bpe_char.model')
@require_sentencepiece
@require_tokenizers
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Tuple = SpeechTaTokenizer
__lowercase: int = False
__lowercase: List[str] = True
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
snake_case_ = SpeechTaTokenizer(UpperCAmelCase_ )
snake_case_ = AddedToken("""<mask>""" , lstrip=UpperCAmelCase_ , rstrip=UpperCAmelCase_ )
snake_case_ = mask_token
tokenizer.add_special_tokens({"""mask_token""": mask_token} )
tokenizer.add_tokens(["""<ctc_blank>"""] )
tokenizer.save_pretrained(self.tmpdirname )
def lowerCAmelCase ( self : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = """this is a test"""
snake_case_ = """this is a test"""
return input_text, output_text
def lowerCAmelCase ( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : Any=False , UpperCAmelCase_ : Tuple=20 , UpperCAmelCase_ : Dict=5 ) ->List[Any]:
"""simple docstring"""
snake_case_ , snake_case_ = self.get_input_output_texts(UpperCAmelCase_ )
snake_case_ = tokenizer.encode(UpperCAmelCase_ , add_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.decode(UpperCAmelCase_ , clean_up_tokenization_spaces=UpperCAmelCase_ )
return text, ids
def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = """<pad>"""
snake_case_ = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(UpperCAmelCase_ ) , UpperCAmelCase_ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(UpperCAmelCase_ ) , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-4] , """œ""" )
self.assertEqual(vocab_keys[-2] , """<mask>""" )
self.assertEqual(vocab_keys[-1] , """<ctc_blank>""" )
self.assertEqual(len(UpperCAmelCase_ ) , 81 )
def lowerCAmelCase ( self : Optional[int] ) ->int:
"""simple docstring"""
self.assertEqual(self.get_tokenizer().vocab_size , 79 )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = self.get_tokenizers(do_lower_case=UpperCAmelCase_ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
snake_case_ = ["""aaaaa bbbbbb""", """cccccccccdddddddd"""]
snake_case_ = tokenizer.add_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode("""aaaaa bbbbbb low cccccccccdddddddd l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
snake_case_ = {"""eos_token""": """>>>>|||<||<<|<<""", """pad_token""": """<<<<<|||>|>>>>|>"""}
snake_case_ = tokenizer.add_special_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size_a + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode(
""">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->List[str]:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = tokenizer.tokenize("""This is a test""" )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [SPIECE_UNDERLINE, """T""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """a""", SPIECE_UNDERLINE, """t""", """e""", """s""", """t"""] )
# fmt: on
self.assertListEqual(
tokenizer.convert_tokens_to_ids(UpperCAmelCase_ ) , [4, 32, 11, 10, 12, 4, 10, 12, 4, 7, 4, 6, 5, 12, 6] , )
snake_case_ = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """92000""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
snake_case_ = tokenizer.convert_tokens_to_ids(UpperCAmelCase_ )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [4, 30, 4, 20, 7, 12, 4, 25, 8, 13, 9, 4, 10, 9, 4, 3, 23, 4, 7, 9, 14, 4, 6, 11, 10, 12, 4, 10, 12, 4, 19, 7, 15, 12, 73, 26] )
# fmt: on
snake_case_ = tokenizer.convert_ids_to_tokens(UpperCAmelCase_ )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """<unk>""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = [
"""Transformers (formerly known as pytorch-transformers and pytorch-pretrained-bert) provides """
"""general-purpose architectures (BERT, GPT, RoBERTa, XLM, DistilBert, XLNet...) for Natural """
"""Language Understanding (NLU) and Natural Language Generation (NLG) with over thirty-two pretrained """
"""models in one hundred plus languages and deep interoperability between Jax, PyTorch and TensorFlow.""",
"""BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly """
"""conditioning on both left and right context in all layers.""",
"""The quick brown fox jumps over the lazy dog.""",
]
# fmt: off
snake_case_ = {
"""input_ids""": [
[4, 32, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 64, 19, 8, 13, 18, 5, 13, 15, 22, 4, 28, 9, 8, 20, 9, 4, 7, 12, 4, 24, 22, 6, 8, 13, 17, 11, 39, 6, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 7, 9, 14, 4, 24, 22, 6, 8, 13, 17, 11, 39, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 39, 25, 5, 13, 6, 63, 4, 24, 13, 8, 27, 10, 14, 5, 12, 4, 21, 5, 9, 5, 13, 7, 15, 39, 24, 16, 13, 24, 8, 12, 5, 4, 7, 13, 17, 11, 10, 6, 5, 17, 6, 16, 13, 5, 12, 4, 64, 40, 47, 54, 32, 23, 4, 53, 49, 32, 23, 4, 54, 8, 40, 47, 54, 32, 7, 23, 4, 69, 52, 43, 23, 4, 51, 10, 12, 6, 10, 15, 40, 5, 13, 6, 23, 4, 69, 52, 48, 5, 6, 26, 26, 26, 63, 4, 19, 8, 13, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 61, 9, 14, 5, 13, 12, 6, 7, 9, 14, 10, 9, 21, 4, 64, 48, 52, 61, 63, 4, 7, 9, 14, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 53, 5, 9, 5, 13, 7, 6, 10, 8, 9, 4, 64, 48, 52, 53, 63, 4, 20, 10, 6, 11, 4, 8, 27, 5, 13, 4, 6, 11, 10, 13, 6, 22, 39, 6, 20, 8, 4, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 4, 18, 8, 14, 5, 15, 12, 4, 10, 9, 4, 8, 9, 5, 4, 11, 16, 9, 14, 13, 5, 14, 4, 24, 15, 16, 12, 4, 15, 7, 9, 21, 16, 7, 21, 5, 12, 4, 7, 9, 14, 4, 14, 5, 5, 24, 4, 10, 9, 6, 5, 13, 8, 24, 5, 13, 7, 25, 10, 15, 10, 6, 22, 4, 25, 5, 6, 20, 5, 5, 9, 4, 58, 7, 37, 23, 4, 49, 22, 32, 8, 13, 17, 11, 4, 7, 9, 14, 4, 32, 5, 9, 12, 8, 13, 55, 15, 8, 20, 26, 2],
[4, 40, 47, 54, 32, 4, 10, 12, 4, 14, 5, 12, 10, 21, 9, 5, 14, 4, 6, 8, 4, 24, 13, 5, 39, 6, 13, 7, 10, 9, 4, 14, 5, 5, 24, 4, 25, 10, 14, 10, 13, 5, 17, 6, 10, 8, 9, 7, 15, 4, 13, 5, 24, 13, 5, 12, 5, 9, 6, 7, 6, 10, 8, 9, 12, 4, 19, 13, 8, 18, 4, 16, 9, 15, 7, 25, 5, 15, 5, 14, 4, 6, 5, 37, 6, 4, 25, 22, 4, 46, 8, 10, 9, 6, 15, 22, 4, 17, 8, 9, 14, 10, 6, 10, 8, 9, 10, 9, 21, 4, 8, 9, 4, 25, 8, 6, 11, 4, 15, 5, 19, 6, 4, 7, 9, 14, 4, 13, 10, 21, 11, 6, 4, 17, 8, 9, 6, 5, 37, 6, 4, 10, 9, 4, 7, 15, 15, 4, 15, 7, 22, 5, 13, 12, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[4, 32, 11, 5, 4, 45, 16, 10, 17, 28, 4, 25, 13, 8, 20, 9, 4, 19, 8, 37, 4, 46, 16, 18, 24, 12, 4, 8, 27, 5, 13, 4, 6, 11, 5, 4, 15, 7, 57, 22, 4, 14, 8, 21, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
],
"""attention_mask""": [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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],
]
}
# fmt: on
self.tokenizer_integration_test_util(
expected_encoding=UpperCAmelCase_ , model_name="""microsoft/speecht5_asr""" , revision="""c5ef64c71905caeccde0e4462ef3f9077224c524""" , sequences=UpperCAmelCase_ , )
| 347 | 0 |
'''simple docstring'''
import os
import unicodedata
from shutil import copyfile
from typing import Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import AddedToken, PreTrainedTokenizer
from ...utils import logging
__lowercase : str = logging.get_logger(__name__)
__lowercase : int = {'vocab_file': 'spiece.model'}
__lowercase : int = {
'vocab_file': {
'albert-base-v1': 'https://huggingface.co/albert-base-v1/resolve/main/spiece.model',
'albert-large-v1': 'https://huggingface.co/albert-large-v1/resolve/main/spiece.model',
'albert-xlarge-v1': 'https://huggingface.co/albert-xlarge-v1/resolve/main/spiece.model',
'albert-xxlarge-v1': 'https://huggingface.co/albert-xxlarge-v1/resolve/main/spiece.model',
'albert-base-v2': 'https://huggingface.co/albert-base-v2/resolve/main/spiece.model',
'albert-large-v2': 'https://huggingface.co/albert-large-v2/resolve/main/spiece.model',
'albert-xlarge-v2': 'https://huggingface.co/albert-xlarge-v2/resolve/main/spiece.model',
'albert-xxlarge-v2': 'https://huggingface.co/albert-xxlarge-v2/resolve/main/spiece.model',
}
}
__lowercase : Union[str, Any] = {
'albert-base-v1': 5_12,
'albert-large-v1': 5_12,
'albert-xlarge-v1': 5_12,
'albert-xxlarge-v1': 5_12,
'albert-base-v2': 5_12,
'albert-large-v2': 5_12,
'albert-xlarge-v2': 5_12,
'albert-xxlarge-v2': 5_12,
}
__lowercase : Any = '▁'
class __UpperCamelCase ( lowerCAmelCase_ ):
A_ = VOCAB_FILES_NAMES
A_ = PRETRAINED_VOCAB_FILES_MAP
A_ = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
def __init__( self , __a , __a=True , __a=True , __a=False , __a="[CLS]" , __a="[SEP]" , __a="<unk>" , __a="[SEP]" , __a="<pad>" , __a="[CLS]" , __a="[MASK]" , __a = None , **__a , ):
'''simple docstring'''
__a : List[str] = (
AddedToken(__a , lstrip=__a , rstrip=__a , normalized=__a )
if isinstance(__a , __a )
else mask_token
)
__a : Dict = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
do_lower_case=__a , remove_space=__a , keep_accents=__a , bos_token=__a , eos_token=__a , unk_token=__a , sep_token=__a , pad_token=__a , cls_token=__a , mask_token=__a , sp_model_kwargs=self.sp_model_kwargs , **__a , )
__a : Any = do_lower_case
__a : Any = remove_space
__a : Optional[Any] = keep_accents
__a : Union[str, Any] = vocab_file
__a : Any = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(__a )
@property
def __UpperCAmelCase ( self ):
'''simple docstring'''
return len(self.sp_model )
def __UpperCAmelCase ( self ):
'''simple docstring'''
__a : Optional[int] = {self.convert_ids_to_tokens(__a ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def __getstate__( self ):
'''simple docstring'''
__a : List[Any] = self.__dict__.copy()
__a : List[str] = None
return state
def __setstate__( self , __a ):
'''simple docstring'''
__a : Optional[Any] = d
# for backward compatibility
if not hasattr(self , 'sp_model_kwargs' ):
__a : Optional[Any] = {}
__a : Union[str, Any] = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def __UpperCAmelCase ( self , __a ):
'''simple docstring'''
if self.remove_space:
__a : Union[str, Any] = ' '.join(inputs.strip().split() )
else:
__a : Dict = inputs
__a : Optional[Any] = outputs.replace('``' , '"' ).replace('\'\'' , '"' )
if not self.keep_accents:
__a : Optional[Any] = unicodedata.normalize('NFKD' , __a )
__a : List[str] = ''.join([c for c in outputs if not unicodedata.combining(__a )] )
if self.do_lower_case:
__a : Tuple = outputs.lower()
return outputs
def __UpperCAmelCase ( self , __a ):
'''simple docstring'''
__a : Dict = self.preprocess_text(__a )
__a : Tuple = self.sp_model.encode(__a , out_type=__a )
__a : List[Any] = []
for piece in pieces:
if len(__a ) > 1 and piece[-1] == str(',' ) and piece[-2].isdigit():
__a : List[str] = self.sp_model.EncodeAsPieces(piece[:-1].replace(__a , '' ) )
if piece[0] != SPIECE_UNDERLINE and cur_pieces[0][0] == SPIECE_UNDERLINE:
if len(cur_pieces[0] ) == 1:
__a : int = cur_pieces[1:]
else:
__a : str = cur_pieces[0][1:]
cur_pieces.append(piece[-1] )
new_pieces.extend(__a )
else:
new_pieces.append(__a )
return new_pieces
def __UpperCAmelCase ( self , __a ):
'''simple docstring'''
return self.sp_model.PieceToId(__a )
def __UpperCAmelCase ( self , __a ):
'''simple docstring'''
return self.sp_model.IdToPiece(__a )
def __UpperCAmelCase ( self , __a ):
'''simple docstring'''
__a : Optional[int] = []
__a : List[Any] = ''
__a : int = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(__a ) + token
__a : List[Any] = True
__a : Union[str, Any] = []
else:
current_sub_tokens.append(__a )
__a : Union[str, Any] = False
out_string += self.sp_model.decode(__a )
return out_string.strip()
def __UpperCAmelCase ( self , __a , __a = None ):
'''simple docstring'''
__a : Optional[int] = [self.sep_token_id]
__a : Optional[int] = [self.cls_token_id]
if token_ids_a is None:
return cls + token_ids_a + sep
return cls + token_ids_a + sep + token_ids_a + sep
def __UpperCAmelCase ( self , __a , __a = None , __a = False ):
'''simple docstring'''
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=__a , token_ids_a=__a , already_has_special_tokens=__a )
if token_ids_a is not None:
return [1] + ([0] * len(__a )) + [1] + ([0] * len(__a )) + [1]
return [1] + ([0] * len(__a )) + [1]
def __UpperCAmelCase ( self , __a , __a = None ):
'''simple docstring'''
__a : Dict = [self.sep_token_id]
__a : Dict = [self.cls_token_id]
if token_ids_a is None:
return len(cls + token_ids_a + sep ) * [0]
return len(cls + token_ids_a + sep ) * [0] + len(token_ids_a + sep ) * [1]
def __UpperCAmelCase ( self , __a , __a = None ):
'''simple docstring'''
if not os.path.isdir(__a ):
logger.error(f"""Vocabulary path ({save_directory}) should be a directory""" )
return
__a : Tuple = os.path.join(
__a , (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file'] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(__a ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , __a )
elif not os.path.isfile(self.vocab_file ):
with open(__a , 'wb' ) as fi:
__a : int = self.sp_model.serialized_model_proto()
fi.write(__a )
return (out_vocab_file,)
| 27 |
"""simple docstring"""
import datasets
__SCREAMING_SNAKE_CASE : Tuple = '\\n@InProceedings{conneau2018xnli,\n author = "Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin",\n title = "XNLI: Evaluating Cross-lingual Sentence Representations",\n booktitle = "Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing",\n year = "2018",\n publisher = "Association for Computational Linguistics",\n location = "Brussels, Belgium",\n}\n'
__SCREAMING_SNAKE_CASE : Dict = '\\nXNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n'
__SCREAMING_SNAKE_CASE : List[str] = '\nComputes XNLI score which is just simple accuracy.\nArgs:\n predictions: Predicted labels.\n references: Ground truth labels.\nReturns:\n \'accuracy\': accuracy\nExamples:\n\n >>> predictions = [0, 1]\n >>> references = [0, 1]\n >>> xnli_metric = datasets.load_metric("xnli")\n >>> results = xnli_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {\'accuracy\': 1.0}\n'
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
return (preds == labels).mean()
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION)
class __A (datasets.Metric):
'''simple docstring'''
def lowerCAmelCase ( self : str ) ->Any:
"""simple docstring"""
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
"""predictions""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
"""references""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
} ) , codebase_urls=[] , reference_urls=[] , format="""numpy""" , )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any ) ->int:
"""simple docstring"""
return {"accuracy": simple_accuracy(UpperCAmelCase_ , UpperCAmelCase_ )}
| 347 | 0 |
'''simple docstring'''
from typing import Dict, List, Optional, Tuple, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import (
center_crop,
flip_channel_order,
get_resize_output_image_size,
rescale,
resize,
to_channel_dimension_format,
)
from ...image_utils import (
ChannelDimension,
ImageInput,
PILImageResampling,
make_list_of_images,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_torch_available, is_torch_tensor, is_vision_available, logging
if is_vision_available():
import PIL
if is_torch_available():
import torch
_lowerCamelCase : Dict = logging.get_logger(__name__)
class SCREAMING_SNAKE_CASE ( _a ):
"""simple docstring"""
_SCREAMING_SNAKE_CASE = ["""pixel_values"""]
def __init__( self : Optional[Any] , UpperCamelCase__ : bool = True , UpperCamelCase__ : Dict[str, int] = None , UpperCamelCase__ : PILImageResampling = PILImageResampling.BILINEAR , UpperCamelCase__ : bool = True , UpperCamelCase__ : Union[int, float] = 1 / 2_5_5 , UpperCamelCase__ : bool = True , UpperCamelCase__ : Dict[str, int] = None , UpperCamelCase__ : bool = True , **UpperCamelCase__ : Tuple , ):
"""simple docstring"""
super().__init__(**UpperCamelCase__ )
UpperCamelCase = size if size is not None else {'shortest_edge': 2_2_4}
UpperCamelCase = get_size_dict(UpperCamelCase__ , default_to_square=UpperCamelCase__ )
UpperCamelCase = crop_size if crop_size is not None else {'height': 2_5_6, 'width': 2_5_6}
UpperCamelCase = get_size_dict(UpperCamelCase__ , param_name='crop_size' )
UpperCamelCase = do_resize
UpperCamelCase = size
UpperCamelCase = resample
UpperCamelCase = do_rescale
UpperCamelCase = rescale_factor
UpperCamelCase = do_center_crop
UpperCamelCase = crop_size
UpperCamelCase = do_flip_channel_order
def A ( self : List[Any] , UpperCamelCase__ : np.ndarray , UpperCamelCase__ : Dict[str, int] , UpperCamelCase__ : PILImageResampling = PIL.Image.BILINEAR , UpperCamelCase__ : Optional[Union[str, ChannelDimension]] = None , **UpperCamelCase__ : Union[str, Any] , ):
"""simple docstring"""
UpperCamelCase = get_size_dict(UpperCamelCase__ , default_to_square=UpperCamelCase__ )
if "shortest_edge" not in size:
raise ValueError(f"""The `size` dictionary must contain the key `shortest_edge`. Got {size.keys()}""" )
UpperCamelCase = get_resize_output_image_size(UpperCamelCase__ , size=size['shortest_edge'] , default_to_square=UpperCamelCase__ )
return resize(UpperCamelCase__ , size=UpperCamelCase__ , resample=UpperCamelCase__ , data_format=UpperCamelCase__ , **UpperCamelCase__ )
def A ( self : Dict , UpperCamelCase__ : np.ndarray , UpperCamelCase__ : Dict[str, int] , UpperCamelCase__ : Optional[Union[str, ChannelDimension]] = None , **UpperCamelCase__ : int , ):
"""simple docstring"""
UpperCamelCase = get_size_dict(UpperCamelCase__ )
if "height" not in size or "width" not in size:
raise ValueError(f"""The `size` dictionary must contain the keys `height` and `width`. Got {size.keys()}""" )
return center_crop(UpperCamelCase__ , size=(size['height'], size['width']) , data_format=UpperCamelCase__ , **UpperCamelCase__ )
def A ( self : Optional[int] , UpperCamelCase__ : np.ndarray , UpperCamelCase__ : Union[int, float] , UpperCamelCase__ : Optional[Union[str, ChannelDimension]] = None , **UpperCamelCase__ : Any , ):
"""simple docstring"""
return rescale(UpperCamelCase__ , scale=UpperCamelCase__ , data_format=UpperCamelCase__ , **UpperCamelCase__ )
def A ( self : List[Any] , UpperCamelCase__ : np.ndarray , UpperCamelCase__ : Optional[Union[str, ChannelDimension]] = None ):
"""simple docstring"""
return flip_channel_order(UpperCamelCase__ , data_format=UpperCamelCase__ )
def A ( self : List[Any] , UpperCamelCase__ : ImageInput , UpperCamelCase__ : bool = None , UpperCamelCase__ : Dict[str, int] = None , UpperCamelCase__ : PILImageResampling = None , UpperCamelCase__ : bool = None , UpperCamelCase__ : float = None , UpperCamelCase__ : bool = None , UpperCamelCase__ : Dict[str, int] = None , UpperCamelCase__ : bool = None , UpperCamelCase__ : Optional[Union[str, TensorType]] = None , UpperCamelCase__ : ChannelDimension = ChannelDimension.FIRST , **UpperCamelCase__ : Union[str, Any] , ):
"""simple docstring"""
UpperCamelCase = do_resize if do_resize is not None else self.do_resize
UpperCamelCase = resample if resample is not None else self.resample
UpperCamelCase = do_rescale if do_rescale is not None else self.do_rescale
UpperCamelCase = rescale_factor if rescale_factor is not None else self.rescale_factor
UpperCamelCase = do_center_crop if do_center_crop is not None else self.do_center_crop
UpperCamelCase = (
do_flip_channel_order if do_flip_channel_order is not None else self.do_flip_channel_order
)
UpperCamelCase = size if size is not None else self.size
UpperCamelCase = get_size_dict(UpperCamelCase__ , default_to_square=UpperCamelCase__ )
UpperCamelCase = crop_size if crop_size is not None else self.crop_size
UpperCamelCase = get_size_dict(UpperCamelCase__ , param_name='crop_size' )
UpperCamelCase = make_list_of_images(UpperCamelCase__ )
if not valid_images(UpperCamelCase__ ):
raise ValueError(
'Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, '
'torch.Tensor, tf.Tensor or jax.ndarray.' )
if do_resize and size is None:
raise ValueError('Size must be specified if do_resize is True.' )
if do_rescale and rescale_factor is None:
raise ValueError('Rescale factor must be specified if do_rescale is True.' )
if do_center_crop and crop_size is None:
raise ValueError('Crop size must be specified if do_center_crop is True.' )
# All transformations expect numpy arrays.
UpperCamelCase = [to_numpy_array(UpperCamelCase__ ) for image in images]
if do_resize:
UpperCamelCase = [self.resize(image=UpperCamelCase__ , size=UpperCamelCase__ , resample=UpperCamelCase__ ) for image in images]
if do_center_crop:
UpperCamelCase = [self.center_crop(image=UpperCamelCase__ , size=UpperCamelCase__ ) for image in images]
if do_rescale:
UpperCamelCase = [self.rescale(image=UpperCamelCase__ , scale=UpperCamelCase__ ) for image in images]
# the pretrained checkpoints assume images are BGR, not RGB
if do_flip_channel_order:
UpperCamelCase = [self.flip_channel_order(image=UpperCamelCase__ ) for image in images]
UpperCamelCase = [to_channel_dimension_format(UpperCamelCase__ , UpperCamelCase__ ) for image in images]
UpperCamelCase = {'pixel_values': images}
return BatchFeature(data=UpperCamelCase__ , tensor_type=UpperCamelCase__ )
def A ( self : Optional[Any] , UpperCamelCase__ : List[Any] , UpperCamelCase__ : List[Tuple] = None ):
"""simple docstring"""
UpperCamelCase = outputs.logits
# Resize logits and compute semantic segmentation maps
if target_sizes is not None:
if len(UpperCamelCase__ ) != len(UpperCamelCase__ ):
raise ValueError(
'Make sure that you pass in as many target sizes as the batch dimension of the logits' )
if is_torch_tensor(UpperCamelCase__ ):
UpperCamelCase = target_sizes.numpy()
UpperCamelCase = []
for idx in range(len(UpperCamelCase__ ) ):
UpperCamelCase = torch.nn.functional.interpolate(
logits[idx].unsqueeze(dim=0 ) , size=target_sizes[idx] , mode='bilinear' , align_corners=UpperCamelCase__ )
UpperCamelCase = resized_logits[0].argmax(dim=0 )
semantic_segmentation.append(UpperCamelCase__ )
else:
UpperCamelCase = logits.argmax(dim=1 )
UpperCamelCase = [semantic_segmentation[i] for i in range(semantic_segmentation.shape[0] )]
return semantic_segmentation
| 28 |
"""simple docstring"""
from ..utils import DummyObject, requires_backends
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Tuple ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : int ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Dict ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Union[str, Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[Any] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : str , **UpperCAmelCase_ : int ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : Union[str, Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[Any] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : List[str] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Optional[Any] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Optional[Any] ) ->int:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Dict ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Union[str, Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[int] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Optional[int] ) ->Any:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
| 347 | 0 |
import multiprocessing
import os
from typing import BinaryIO, Optional, Union
import fsspec
from .. import Dataset, Features, NamedSplit, config
from ..formatting import query_table
from ..packaged_modules.json.json import Json
from ..utils import logging
from ..utils.typing import NestedDataStructureLike, PathLike
from .abc import AbstractDatasetReader
class lowerCamelCase (_snake_case ):
'''simple docstring'''
def __init__( self , _UpperCamelCase , _UpperCamelCase = None , _UpperCamelCase = None , _UpperCamelCase = None , _UpperCamelCase = False , _UpperCamelCase = False , _UpperCamelCase = None , _UpperCamelCase = None , **_UpperCamelCase , ) -> Optional[int]:
super().__init__(
_UpperCamelCase , split=_UpperCamelCase , features=_UpperCamelCase , cache_dir=_UpperCamelCase , keep_in_memory=_UpperCamelCase , streaming=_UpperCamelCase , num_proc=_UpperCamelCase , **_UpperCamelCase , )
UpperCAmelCase_ : Any = field
UpperCAmelCase_ : Tuple = path_or_paths if isinstance(_UpperCamelCase , _UpperCamelCase ) else {self.split: path_or_paths}
UpperCAmelCase_ : int = Json(
cache_dir=_UpperCamelCase , data_files=_UpperCamelCase , features=_UpperCamelCase , field=_UpperCamelCase , **_UpperCamelCase , )
def __UpperCAmelCase ( self ) -> List[Any]:
# Build iterable dataset
if self.streaming:
UpperCAmelCase_ : Optional[Any] = self.builder.as_streaming_dataset(split=self.split )
# Build regular (map-style) dataset
else:
UpperCAmelCase_ : Optional[int] = None
UpperCAmelCase_ : List[Any] = None
UpperCAmelCase_ : Any = None
UpperCAmelCase_ : Any = None
self.builder.download_and_prepare(
download_config=_UpperCamelCase , download_mode=_UpperCamelCase , verification_mode=_UpperCamelCase , base_path=_UpperCamelCase , num_proc=self.num_proc , )
UpperCAmelCase_ : List[str] = self.builder.as_dataset(
split=self.split , verification_mode=_UpperCamelCase , in_memory=self.keep_in_memory )
return dataset
class lowerCamelCase :
'''simple docstring'''
def __init__( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase = None , _UpperCamelCase = None , **_UpperCamelCase , ) -> Union[str, Any]:
if num_proc is not None and num_proc <= 0:
raise ValueError(f"num_proc {num_proc} must be an integer > 0." )
UpperCAmelCase_ : List[Any] = dataset
UpperCAmelCase_ : Optional[int] = path_or_buf
UpperCAmelCase_ : List[str] = batch_size if batch_size else config.DEFAULT_MAX_BATCH_SIZE
UpperCAmelCase_ : str = num_proc
UpperCAmelCase_ : Optional[int] = 'utf-8'
UpperCAmelCase_ : Optional[int] = to_json_kwargs
def __UpperCAmelCase ( self ) -> int:
UpperCAmelCase_ : Dict = self.to_json_kwargs.pop('path_or_buf' , _UpperCamelCase )
UpperCAmelCase_ : Union[str, Any] = self.to_json_kwargs.pop('orient' , 'records' )
UpperCAmelCase_ : Any = self.to_json_kwargs.pop('lines' , True if orient == 'records' else False )
UpperCAmelCase_ : str = self.to_json_kwargs.pop('index' , False if orient in ['split', 'table'] else True )
UpperCAmelCase_ : List[str] = self.to_json_kwargs.pop('compression' , _UpperCamelCase )
if compression not in [None, "infer", "gzip", "bz2", "xz"]:
raise NotImplementedError(f"`datasets` currently does not support {compression} compression" )
if isinstance(self.path_or_buf , (str, bytes, os.PathLike) ):
with fsspec.open(self.path_or_buf , 'wb' , compression=_UpperCamelCase ) as buffer:
UpperCAmelCase_ : Any = self._write(file_obj=_UpperCamelCase , orient=_UpperCamelCase , lines=_UpperCamelCase , index=_UpperCamelCase , **self.to_json_kwargs )
else:
if compression:
raise NotImplementedError(
f"The compression parameter is not supported when writing to a buffer, but compression={compression}"
' was passed. Please provide a local path instead.' )
UpperCAmelCase_ : Tuple = self._write(
file_obj=self.path_or_buf , orient=_UpperCamelCase , lines=_UpperCamelCase , index=_UpperCamelCase , **self.to_json_kwargs )
return written
def __UpperCAmelCase ( self , _UpperCamelCase ) -> List[Any]:
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ : List[str] = args
UpperCAmelCase_ : List[str] = query_table(
table=self.dataset.data , key=slice(_UpperCamelCase , offset + self.batch_size ) , indices=self.dataset._indices , )
UpperCAmelCase_ : str = batch.to_pandas().to_json(
path_or_buf=_UpperCamelCase , orient=_UpperCamelCase , lines=_UpperCamelCase , index=_UpperCamelCase , **_UpperCamelCase )
if not json_str.endswith('\n' ):
json_str += "\n"
return json_str.encode(self.encoding )
def __UpperCAmelCase ( self , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase , _UpperCamelCase , **_UpperCamelCase , ) -> int:
UpperCAmelCase_ : Any = 0
if self.num_proc is None or self.num_proc == 1:
for offset in logging.tqdm(
range(0 , len(self.dataset ) , self.batch_size ) , unit='ba' , disable=not logging.is_progress_bar_enabled() , desc='Creating json from Arrow format' , ):
UpperCAmelCase_ : List[Any] = self._batch_json((offset, orient, lines, index, to_json_kwargs) )
written += file_obj.write(_UpperCamelCase )
else:
UpperCAmelCase_ , UpperCAmelCase_ : List[Any] = len(self.dataset ), self.batch_size
with multiprocessing.Pool(self.num_proc ) as pool:
for json_str in logging.tqdm(
pool.imap(
self._batch_json , [(offset, orient, lines, index, to_json_kwargs) for offset in range(0 , _UpperCamelCase , _UpperCamelCase )] , ) , total=(num_rows // batch_size) + 1 if num_rows % batch_size else num_rows // batch_size , unit='ba' , disable=not logging.is_progress_bar_enabled() , desc='Creating json from Arrow format' , ):
written += file_obj.write(_UpperCamelCase )
return written
| 29 |
"""simple docstring"""
import warnings
from ...utils import logging
from .image_processing_mobilevit import MobileViTImageProcessor
__SCREAMING_SNAKE_CASE : int = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
def __init__( self : str , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->None:
"""simple docstring"""
warnings.warn(
"""The class MobileViTFeatureExtractor is deprecated and will be removed in version 5 of Transformers."""
""" Please use MobileViTImageProcessor instead.""" , UpperCAmelCase_ , )
super().__init__(*UpperCAmelCase_ , **UpperCAmelCase_ )
| 347 | 0 |
import json
import os
import subprocess
import unittest
from ast import literal_eval
import pytest
from parameterized import parameterized, parameterized_class
from . import is_sagemaker_available
if is_sagemaker_available():
from sagemaker import Session, TrainingJobAnalytics
from sagemaker.huggingface import HuggingFace
@pytest.mark.skipif(
literal_eval(os.getenv('TEST_SAGEMAKER' , 'False' ) ) is not True , reason='Skipping test because should only be run when releasing minor transformers version' , )
@pytest.mark.usefixtures('sm_env' )
@parameterized_class(
[
{
'framework': 'pytorch',
'script': 'run_glue_model_parallelism.py',
'model_name_or_path': 'roberta-large',
'instance_type': 'ml.p3dn.24xlarge',
'results': {'train_runtime': 1_600, 'eval_accuracy': 0.3, 'eval_loss': 1.2},
},
{
'framework': 'pytorch',
'script': 'run_glue.py',
'model_name_or_path': 'roberta-large',
'instance_type': 'ml.p3dn.24xlarge',
'results': {'train_runtime': 1_600, 'eval_accuracy': 0.3, 'eval_loss': 1.2},
},
] )
class lowercase__( unittest.TestCase ):
"""simple docstring"""
def _lowercase ( self : str ) -> str:
if self.framework == "pytorch":
subprocess.run(
f'''cp ./examples/pytorch/text-classification/run_glue.py {self.env.test_path}/run_glue.py'''.split() , encoding='''utf-8''' , check=SCREAMING_SNAKE_CASE_ , )
assert hasattr(self , '''env''' )
def _lowercase ( self : int , SCREAMING_SNAKE_CASE_ : Optional[int] ) -> Optional[int]:
# configuration for running training on smdistributed Model Parallel
lowercase_ = {
'''enabled''': True,
'''processes_per_host''': 8,
}
lowercase_ = {
'''enabled''': True,
'''parameters''': {
'''microbatches''': 4,
'''placement_strategy''': '''spread''',
'''pipeline''': '''interleaved''',
'''optimize''': '''speed''',
'''partitions''': 4,
'''ddp''': True,
},
}
lowercase_ = {'''smdistributed''': {'''modelparallel''': smp_options}, '''mpi''': mpi_options}
lowercase_ = '''trainer''' if self.script == '''run_glue.py''' else '''smtrainer'''
# creates estimator
return HuggingFace(
entry_point=self.script , source_dir=self.env.test_path , role=self.env.role , image_uri=self.env.image_uri , base_job_name=f'''{self.env.base_job_name}-{instance_count}-smp-{name_extension}''' , instance_count=SCREAMING_SNAKE_CASE_ , instance_type=self.instance_type , debugger_hook_config=SCREAMING_SNAKE_CASE_ , hyperparameters={
**self.env.hyperparameters,
'''model_name_or_path''': self.model_name_or_path,
'''max_steps''': 5_0_0,
} , metric_definitions=self.env.metric_definitions , distribution=SCREAMING_SNAKE_CASE_ , py_version='''py36''' , )
def _lowercase ( self : int , SCREAMING_SNAKE_CASE_ : Tuple ) -> List[str]:
TrainingJobAnalytics(SCREAMING_SNAKE_CASE_ ).export_csv(f'''{self.env.test_path}/{job_name}_metrics.csv''' )
@parameterized.expand([(1,)] )
def _lowercase ( self : int , SCREAMING_SNAKE_CASE_ : Any ) -> Any:
# create estimator
lowercase_ = self.create_estimator(SCREAMING_SNAKE_CASE_ )
# run training
estimator.fit()
# result dataframe
lowercase_ = TrainingJobAnalytics(estimator.latest_training_job.name ).dataframe()
# extract kpis
lowercase_ = list(result_metrics_df[result_metrics_df.metric_name == '''eval_accuracy''']['''value'''] )
lowercase_ = list(result_metrics_df[result_metrics_df.metric_name == '''eval_loss''']['''value'''] )
# get train time from SageMaker job, this includes starting, preprocessing, stopping
lowercase_ = (
Session().describe_training_job(estimator.latest_training_job.name ).get('''TrainingTimeInSeconds''' , 9_9_9_9_9_9 )
)
# assert kpis
assert train_runtime <= self.results["train_runtime"]
assert all(t >= self.results['''eval_accuracy'''] for t in eval_accuracy )
assert all(t <= self.results['''eval_loss'''] for t in eval_loss )
# dump tests result into json file to share in PR
with open(f'''{estimator.latest_training_job.name}.json''' , '''w''' ) as outfile:
json.dump({'''train_time''': train_runtime, '''eval_accuracy''': eval_accuracy, '''eval_loss''': eval_loss} , SCREAMING_SNAKE_CASE_ )
| 30 |
"""simple docstring"""
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Any:
snake_case_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
("""cls_token""", """vit.embeddings.cls_token"""),
("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight"""),
("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias"""),
("""pos_embed""", """vit.embeddings.position_embeddings"""),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("""norm.weight""", """layernorm.weight"""),
("""norm.bias""", """layernorm.bias"""),
("""pre_logits.fc.weight""", """pooler.dense.weight"""),
("""pre_logits.fc.bias""", """pooler.dense.bias"""),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
snake_case_ = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("""norm.weight""", """vit.layernorm.weight"""),
("""norm.bias""", """vit.layernorm.bias"""),
("""head.weight""", """classifier.weight"""),
("""head.bias""", """classifier.bias"""),
] )
return rename_keys
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
snake_case_ = """"""
else:
snake_case_ = """vit."""
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case_ = in_proj_weight[
: config.hidden_size, :
]
snake_case_ = in_proj_bias[: config.hidden_size]
snake_case_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case_ = in_proj_weight[
-config.hidden_size :, :
]
snake_case_ = in_proj_bias[-config.hidden_size :]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = ["""head.weight""", """head.bias"""]
for k in ignore_keys:
state_dict.pop(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> str:
snake_case_ = dct.pop(_SCREAMING_SNAKE_CASE )
snake_case_ = val
def _a ( ) -> Any:
snake_case_ = """http://images.cocodataset.org/val2017/000000039769.jpg"""
snake_case_ = Image.open(requests.get(_SCREAMING_SNAKE_CASE , stream=_SCREAMING_SNAKE_CASE ).raw )
return im
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = ViTConfig()
snake_case_ = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
snake_case_ = True
snake_case_ = int(vit_name[-12:-10] )
snake_case_ = int(vit_name[-9:-6] )
else:
snake_case_ = 1_000
snake_case_ = """huggingface/label-files"""
snake_case_ = """imagenet-1k-id2label.json"""
snake_case_ = json.load(open(hf_hub_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , repo_type="""dataset""" ) , """r""" ) )
snake_case_ = {int(_SCREAMING_SNAKE_CASE ): v for k, v in idalabel.items()}
snake_case_ = idalabel
snake_case_ = {v: k for k, v in idalabel.items()}
snake_case_ = int(vit_name[-6:-4] )
snake_case_ = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith("""tiny""" ):
snake_case_ = 192
snake_case_ = 768
snake_case_ = 12
snake_case_ = 3
elif vit_name[9:].startswith("""small""" ):
snake_case_ = 384
snake_case_ = 1_536
snake_case_ = 12
snake_case_ = 6
else:
pass
else:
if vit_name[4:].startswith("""small""" ):
snake_case_ = 768
snake_case_ = 2_304
snake_case_ = 8
snake_case_ = 8
elif vit_name[4:].startswith("""base""" ):
pass
elif vit_name[4:].startswith("""large""" ):
snake_case_ = 1_024
snake_case_ = 4_096
snake_case_ = 24
snake_case_ = 16
elif vit_name[4:].startswith("""huge""" ):
snake_case_ = 1_280
snake_case_ = 5_120
snake_case_ = 32
snake_case_ = 16
# load original model from timm
snake_case_ = timm.create_model(_SCREAMING_SNAKE_CASE , pretrained=_SCREAMING_SNAKE_CASE )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case_ = timm_model.state_dict()
if base_model:
remove_classification_head_(_SCREAMING_SNAKE_CASE )
snake_case_ = create_rename_keys(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
for src, dest in rename_keys:
rename_key(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
read_in_q_k_v(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# load HuggingFace model
if vit_name[-5:] == "in21k":
snake_case_ = ViTModel(_SCREAMING_SNAKE_CASE ).eval()
else:
snake_case_ = ViTForImageClassification(_SCREAMING_SNAKE_CASE ).eval()
model.load_state_dict(_SCREAMING_SNAKE_CASE )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
snake_case_ = DeiTImageProcessor(size=config.image_size )
else:
snake_case_ = ViTImageProcessor(size=config.image_size )
snake_case_ = image_processor(images=prepare_img() , return_tensors="""pt""" )
snake_case_ = encoding["""pixel_values"""]
snake_case_ = model(_SCREAMING_SNAKE_CASE )
if base_model:
snake_case_ = timm_model.forward_features(_SCREAMING_SNAKE_CASE )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.pooler_output , atol=1E-3 )
else:
snake_case_ = timm_model(_SCREAMING_SNAKE_CASE )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.logits , atol=1E-3 )
Path(_SCREAMING_SNAKE_CASE ).mkdir(exist_ok=_SCREAMING_SNAKE_CASE )
print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_SCREAMING_SNAKE_CASE )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--vit_name',
default='vit_base_patch16_224',
type=str,
help='Name of the ViT timm model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
__SCREAMING_SNAKE_CASE : int = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 347 | 0 |
'''simple docstring'''
# Function to print upper half of diamond (pyramid)
def UpperCamelCase_ ( _UpperCAmelCase : Tuple ) -> Dict:
"""simple docstring"""
for i in range(0 , _UpperCAmelCase ):
for _ in range(0 , n - i - 1 ): # printing spaces
print(" " , end="" )
for _ in range(0 , i + 1 ): # printing stars
print("* " , end="" )
print()
def UpperCamelCase_ ( _UpperCAmelCase : Optional[Any] ) -> List[Any]:
"""simple docstring"""
for i in range(_UpperCAmelCase , 0 , -1 ):
for _ in range(_UpperCAmelCase , 0 , -1 ): # printing stars
print("* " , end="" )
print()
for _ in range(n - i + 1 , 0 , -1 ): # printing spaces
print(" " , end="" )
def UpperCamelCase_ ( _UpperCAmelCase : Tuple ) -> Dict:
"""simple docstring"""
if n <= 0:
print(" ... .... nothing printing :(" )
return
floyd(_UpperCAmelCase ) # upper half
reverse_floyd(_UpperCAmelCase ) # lower half
if __name__ == "__main__":
print(R"""| /\ | |- | |- |--| |\ /| |-""")
print(R"""|/ \| |- |_ |_ |__| | \/ | |_""")
__SCREAMING_SNAKE_CASE : str = 1
while K:
__SCREAMING_SNAKE_CASE : Tuple = int(input("""enter the number and , and see the magic : """))
print()
pretty_print(user_number)
__SCREAMING_SNAKE_CASE : Any = int(input("""press 0 to exit... and 1 to continue..."""))
print("""Good Bye...""")
| 31 |
"""simple docstring"""
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A (unittest.TestCase):
'''simple docstring'''
def __init__( self : List[Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Tuple=13 , UpperCAmelCase_ : List[Any]=7 , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=99 , UpperCAmelCase_ : str=32 , UpperCAmelCase_ : Tuple=5 , UpperCAmelCase_ : Union[str, Any]=4 , UpperCAmelCase_ : Any=37 , UpperCAmelCase_ : int="gelu" , UpperCAmelCase_ : Any=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Dict=512 , UpperCAmelCase_ : Optional[Any]=16 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : str=0.02 , UpperCAmelCase_ : str=4 , ) ->Tuple:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_attention_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = type_vocab_size
snake_case_ = type_sequence_label_size
snake_case_ = initializer_range
snake_case_ = num_choices
def lowerCAmelCase ( self : Optional[int] ) ->str:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_attention_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = None
if self.use_token_type_ids:
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=UpperCAmelCase_ , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def lowerCAmelCase ( self : List[str] ) ->Dict:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """token_type_ids""": token_type_ids, """attention_mask""": attention_mask}
return config, inputs_dict
@require_flax
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Union[str, Any] = True
__lowercase: int = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = FlaxRoFormerModelTester(self )
@slow
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""junnyu/roformer_chinese_small""" , from_pt=UpperCAmelCase_ )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
@require_flax
class __A (unittest.TestCase):
'''simple docstring'''
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = FlaxRoFormerForMaskedLM.from_pretrained("""junnyu/roformer_chinese_base""" )
snake_case_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
snake_case_ = model(UpperCAmelCase_ )[0]
snake_case_ = 50_000
snake_case_ = (1, 6, vocab_size)
self.assertEqual(output.shape , UpperCAmelCase_ )
snake_case_ = jnp.array(
[[[-0.1_205, -1.0_265, 0.2_922], [-1.5_134, 0.1_974, 0.1_519], [-5.0_135, -3.9_003, -0.8_404]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=1E-4 ) )
| 347 | 0 |
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
from ..auto import CONFIG_MAPPING
UpperCAmelCase_ : List[str] = logging.get_logger(__name__)
UpperCAmelCase_ : Optional[Any] = {
'microsoft/table-transformer-detection': (
'https://huggingface.co/microsoft/table-transformer-detection/resolve/main/config.json'
),
}
class SCREAMING_SNAKE_CASE__ ( lowercase__ ):
snake_case__ : int = '''table-transformer'''
snake_case__ : Union[str, Any] = ['''past_key_values''']
snake_case__ : Optional[Any] = {
'''hidden_size''': '''d_model''',
'''num_attention_heads''': '''encoder_attention_heads''',
}
def __init__( self : Optional[Any] , SCREAMING_SNAKE_CASE__ : List[Any]=True , SCREAMING_SNAKE_CASE__ : str=None , SCREAMING_SNAKE_CASE__ : Union[str, Any]=3 , SCREAMING_SNAKE_CASE__ : str=1_0_0 , SCREAMING_SNAKE_CASE__ : Optional[Any]=6 , SCREAMING_SNAKE_CASE__ : List[str]=2_0_4_8 , SCREAMING_SNAKE_CASE__ : List[str]=8 , SCREAMING_SNAKE_CASE__ : Dict=6 , SCREAMING_SNAKE_CASE__ : Tuple=2_0_4_8 , SCREAMING_SNAKE_CASE__ : List[str]=8 , SCREAMING_SNAKE_CASE__ : List[str]=0.0 , SCREAMING_SNAKE_CASE__ : List[str]=0.0 , SCREAMING_SNAKE_CASE__ : Any=True , SCREAMING_SNAKE_CASE__ : Optional[Any]="relu" , SCREAMING_SNAKE_CASE__ : Optional[Any]=2_5_6 , SCREAMING_SNAKE_CASE__ : Dict=0.1 , SCREAMING_SNAKE_CASE__ : Optional[int]=0.0 , SCREAMING_SNAKE_CASE__ : Optional[int]=0.0 , SCREAMING_SNAKE_CASE__ : List[str]=0.02 , SCREAMING_SNAKE_CASE__ : Tuple=1.0 , SCREAMING_SNAKE_CASE__ : Any=False , SCREAMING_SNAKE_CASE__ : List[Any]="sine" , SCREAMING_SNAKE_CASE__ : Dict="resnet50" , SCREAMING_SNAKE_CASE__ : Optional[Any]=True , SCREAMING_SNAKE_CASE__ : List[Any]=False , SCREAMING_SNAKE_CASE__ : Dict=1 , SCREAMING_SNAKE_CASE__ : int=5 , SCREAMING_SNAKE_CASE__ : Dict=2 , SCREAMING_SNAKE_CASE__ : Optional[int]=1 , SCREAMING_SNAKE_CASE__ : int=1 , SCREAMING_SNAKE_CASE__ : Any=5 , SCREAMING_SNAKE_CASE__ : List[str]=2 , SCREAMING_SNAKE_CASE__ : Union[str, Any]=0.1 , **SCREAMING_SNAKE_CASE__ : int , ) -> Tuple:
if backbone_config is not None and use_timm_backbone:
raise ValueError('You can\'t specify both `backbone_config` and `use_timm_backbone`.' )
if not use_timm_backbone:
if backbone_config is None:
logger.info('`backbone_config` is `None`. Initializing the config with the default `ResNet` backbone.' )
a_ : Dict = CONFIG_MAPPING['resnet'](out_features=['stage4'] )
elif isinstance(SCREAMING_SNAKE_CASE__ , SCREAMING_SNAKE_CASE__ ):
a_ : int = backbone_config.get('model_type' )
a_ : Any = CONFIG_MAPPING[backbone_model_type]
a_ : Optional[int] = config_class.from_dict(SCREAMING_SNAKE_CASE__ )
# set timm attributes to None
a_ , a_ , a_ : Optional[int] = None, None, None
a_ : Union[str, Any] = use_timm_backbone
a_ : str = backbone_config
a_ : str = num_channels
a_ : Tuple = num_queries
a_ : int = d_model
a_ : Optional[int] = encoder_ffn_dim
a_ : str = encoder_layers
a_ : Optional[Any] = encoder_attention_heads
a_ : Any = decoder_ffn_dim
a_ : int = decoder_layers
a_ : Dict = decoder_attention_heads
a_ : Union[str, Any] = dropout
a_ : Dict = attention_dropout
a_ : Dict = activation_dropout
a_ : Optional[Any] = activation_function
a_ : List[str] = init_std
a_ : Any = init_xavier_std
a_ : Union[str, Any] = encoder_layerdrop
a_ : Optional[int] = decoder_layerdrop
a_ : Union[str, Any] = encoder_layers
a_ : Dict = auxiliary_loss
a_ : str = position_embedding_type
a_ : Union[str, Any] = backbone
a_ : Any = use_pretrained_backbone
a_ : List[str] = dilation
# Hungarian matcher
a_ : Tuple = class_cost
a_ : Optional[Any] = bbox_cost
a_ : Any = giou_cost
# Loss coefficients
a_ : Optional[Any] = mask_loss_coefficient
a_ : Union[str, Any] = dice_loss_coefficient
a_ : Dict = bbox_loss_coefficient
a_ : List[Any] = giou_loss_coefficient
a_ : str = eos_coefficient
super().__init__(is_encoder_decoder=SCREAMING_SNAKE_CASE__ , **SCREAMING_SNAKE_CASE__ )
@property
def SCREAMING_SNAKE_CASE ( self : str ) -> int:
return self.encoder_attention_heads
@property
def SCREAMING_SNAKE_CASE ( self : str ) -> int:
return self.d_model
class SCREAMING_SNAKE_CASE__ ( lowercase__ ):
snake_case__ : int = version.parse('''1.11''' )
@property
def SCREAMING_SNAKE_CASE ( self : Optional[Any] ) -> Mapping[str, Mapping[int, str]]:
return OrderedDict(
[
('pixel_values', {0: 'batch', 1: 'num_channels', 2: 'height', 3: 'width'}),
('pixel_mask', {0: 'batch'}),
] )
@property
def SCREAMING_SNAKE_CASE ( self : Any ) -> float:
return 1E-5
@property
def SCREAMING_SNAKE_CASE ( self : int ) -> int:
return 1_2
| 32 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> bool:
snake_case_ = get_failure_array(_SCREAMING_SNAKE_CASE )
# 2) Step through text searching for pattern
snake_case_ , snake_case_ = 0, 0 # index into text, pattern
while i < len(_SCREAMING_SNAKE_CASE ):
if pattern[j] == text[i]:
if j == (len(_SCREAMING_SNAKE_CASE ) - 1):
return True
j += 1
# if this is a prefix in our pattern
# just go back far enough to continue
elif j > 0:
snake_case_ = failure[j - 1]
continue
i += 1
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> list[int]:
snake_case_ = [0]
snake_case_ = 0
snake_case_ = 1
while j < len(_SCREAMING_SNAKE_CASE ):
if pattern[i] == pattern[j]:
i += 1
elif i > 0:
snake_case_ = failure[i - 1]
continue
j += 1
failure.append(_SCREAMING_SNAKE_CASE )
return failure
if __name__ == "__main__":
# Test 1)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abc1abc12'
__SCREAMING_SNAKE_CASE : Optional[int] = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
__SCREAMING_SNAKE_CASE : List[str] = 'alskfjaldsk23adsfabcabc'
assert kmp(pattern, texta) and not kmp(pattern, texta)
# Test 2)
__SCREAMING_SNAKE_CASE : int = 'ABABX'
__SCREAMING_SNAKE_CASE : Optional[Any] = 'ABABZABABYABABX'
assert kmp(pattern, text)
# Test 3)
__SCREAMING_SNAKE_CASE : Any = 'AAAB'
__SCREAMING_SNAKE_CASE : List[Any] = 'ABAAAAAB'
assert kmp(pattern, text)
# Test 4)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abcdabcy'
__SCREAMING_SNAKE_CASE : str = 'abcxabcdabxabcdabcdabcy'
assert kmp(pattern, text)
# Test 5)
__SCREAMING_SNAKE_CASE : Any = 'aabaabaaa'
assert get_failure_array(pattern) == [0, 1, 0, 1, 2, 3, 4, 5, 2]
| 347 | 0 |
"""simple docstring"""
import argparse
import json
import os
import re
import shutil
import torch
from transformers import BioGptConfig, BioGptForCausalLM
from transformers.models.biogpt.tokenization_biogpt import VOCAB_FILES_NAMES
from transformers.tokenization_utils_base import TOKENIZER_CONFIG_FILE
from transformers.utils import WEIGHTS_NAME, logging
logging.set_verbosity_warning()
__A : List[Any] = 2
class _UpperCAmelCase :
def __init__( self : Tuple , *, # begin keyword-only arguments
A : Tuple="<s>" , A : List[str]="<pad>" , A : Optional[Any]="</s>" , A : str="<unk>" , A : int=None , ) -> Union[str, Any]:
lowercase_ , lowercase_ , lowercase_ , lowercase_ : List[str] = bos, unk, pad, eos
lowercase_ : Tuple = []
lowercase_ : Union[str, Any] = []
lowercase_ : Dict = {}
lowercase_ : List[Any] = self.add_symbol(A )
lowercase_ : Optional[Any] = self.add_symbol(A )
lowercase_ : Optional[Any] = self.add_symbol(A )
lowercase_ : str = self.add_symbol(A )
if extra_special_symbols:
for s in extra_special_symbols:
self.add_symbol(A )
lowercase_ : int = len(self.symbols )
def __eq__( self : str , A : Tuple ) -> Any:
return self.indices == other.indices
def __getitem__( self : int , A : Tuple ) -> Any:
if idx < len(self.symbols ):
return self.symbols[idx]
return self.unk_word
def __len__( self : Any ) -> Union[str, Any]:
return len(self.symbols )
def __contains__( self : Optional[Any] , A : Optional[int] ) -> Dict:
return sym in self.indices
@classmethod
def A ( cls : Optional[int] , A : Dict ) -> Any:
lowercase_ : Any = cls()
d.add_from_file(A )
return d
def A ( self : List[Any] , A : int , A : List[Any]=1 , A : List[str]=False ) -> Dict:
if word in self.indices and not overwrite:
lowercase_ : Optional[int] = self.indices[word]
lowercase_ : Tuple = self.count[idx] + n
return idx
else:
lowercase_ : Dict = len(self.symbols )
lowercase_ : int = idx
self.symbols.append(A )
self.count.append(A )
return idx
def A ( self : int , A : Tuple ) -> List[str]:
return 0
def A ( self : str , A : str ) -> Tuple:
if isinstance(A , A ):
try:
with open(A , '''r''' , encoding='''utf-8''' ) as fd:
self.add_from_file(A )
except FileNotFoundError as fnfe:
raise fnfe
except UnicodeError:
raise Exception('''Incorrect encoding detected in {}, please rebuild the dataset'''.format(A ) )
return
lowercase_ : Any = f.readlines()
lowercase_ : int = self._load_meta(A )
for line in lines[indices_start_line:]:
try:
lowercase_ , lowercase_ : Any = line.rstrip().rsplit(''' ''' , 1 )
if field == "#fairseq:overwrite":
lowercase_ : str = True
lowercase_ , lowercase_ : Union[str, Any] = line.rsplit(''' ''' , 1 )
else:
lowercase_ : Tuple = False
lowercase_ : Optional[int] = int(A )
lowercase_ : Optional[int] = line
if word in self and not overwrite:
raise RuntimeError(
'''Duplicate word found when loading Dictionary: \'{}\'. '''
'''Duplicate words can overwrite earlier ones by adding the '''
'''#fairseq:overwrite flag at the end of the corresponding row '''
'''in the dictionary file. If using the Camembert model, please '''
'''download an updated copy of the model file.'''.format(A ) )
self.add_symbol(A , n=A , overwrite=A )
except ValueError:
raise ValueError('''Incorrect dictionary format, expected \'<token> <cnt> [flags]\'''' )
def lowercase ( __snake_case : Dict ):
# (1) remove word breaking symbol, (2) add word ending symbol where the word is not broken up,
# e.g.: d = {'le@@': 5, 'tt@@': 6, 'er': 7} => {'le': 5, 'tt': 6, 'er</w>': 7}
lowercase_ : Dict = dict((re.sub(r'''@@$''' , '''''' , __snake_case ), v) if k.endswith('''@@''' ) else (re.sub(r'''$''' , '''</w>''' , __snake_case ), v) for k, v in d.items() )
lowercase_ : int = '''<s> <pad> </s> <unk>'''.split()
# restore the special tokens
for k in keep_keys:
del da[F'''{k}</w>''']
lowercase_ : Union[str, Any] = d[k] # restore
return da
def lowercase ( __snake_case : Tuple , __snake_case : Any ):
# prep
if not os.path.exists(__snake_case ):
raise ValueError(F'''path {biogpt_checkpoint_path} does not exist!''' )
os.makedirs(__snake_case , exist_ok=__snake_case )
print(F'''Writing results to {pytorch_dump_folder_path}''' )
# handle various types of models
lowercase_ : Optional[Any] = os.path.join(__snake_case , '''checkpoint.pt''' )
if not os.path.isfile(__snake_case ):
raise ValueError(F'''path to the file {checkpoint_file} does not exist!''' )
lowercase_ : Union[str, Any] = torch.load(__snake_case , map_location='''cpu''' )
lowercase_ : int = chkpt['''cfg''']['''model''']
# dicts
lowercase_ : int = os.path.join(__snake_case , '''dict.txt''' )
if not os.path.isfile(__snake_case ):
raise ValueError(F'''path to the file {dict_file} does not exist!''' )
lowercase_ : str = Dictionary.load(__snake_case )
lowercase_ : List[str] = rewrite_dict_keys(src_dict.indices )
lowercase_ : Dict = len(__snake_case )
lowercase_ : int = os.path.join(__snake_case , VOCAB_FILES_NAMES['''vocab_file'''] )
print(F'''Generating {src_vocab_file} of {src_vocab_size} records''' )
with open(__snake_case , '''w''' , encoding='''utf-8''' ) as f:
f.write(json.dumps(__snake_case , ensure_ascii=__snake_case , indent=__snake_case ) )
# merges_file (bpecodes)
lowercase_ : Optional[int] = os.path.join(__snake_case , '''bpecodes''' )
if not os.path.isfile(__snake_case ):
raise ValueError(F'''path to the file {bpecodes_file} does not exist!''' )
lowercase_ : List[Any] = os.path.join(__snake_case , VOCAB_FILES_NAMES['''merges_file'''] )
shutil.copyfile(__snake_case , __snake_case )
# model config
lowercase_ : Union[str, Any] = os.path.join(__snake_case , '''config.json''' )
lowercase_ : Dict = {
'''activation_dropout''': args['''activation_dropout'''],
'''architectures''': ['''BioGptForCausalLM'''],
'''attention_probs_dropout_prob''': args['''attention_dropout'''],
'''bos_token_id''': 0,
'''eos_token_id''': 2,
'''hidden_act''': args['''activation_fn'''],
'''hidden_dropout_prob''': args['''dropout'''],
'''hidden_size''': args['''decoder_embed_dim'''],
'''initializer_range''': 0.02,
'''intermediate_size''': args['''decoder_ffn_embed_dim'''],
'''layer_norm_eps''': 1e-12,
'''layerdrop''': args['''decoder_layerdrop'''],
'''max_position_embeddings''': args['''max_target_positions'''],
'''model_type''': '''biogpt''',
'''num_attention_heads''': args['''decoder_attention_heads'''],
'''num_hidden_layers''': args['''decoder_layers'''],
'''pad_token_id''': 1,
'''scale_embedding''': not args['''no_scale_embedding'''],
'''tie_word_embeddings''': args['''share_decoder_input_output_embed'''],
'''vocab_size''': src_vocab_size,
}
# good hparam defaults to start with
print(F'''Generating {biogpt_model_config_file}''' )
with open(__snake_case , '''w''' , encoding='''utf-8''' ) as f:
f.write(json.dumps(__snake_case , ensure_ascii=__snake_case , indent=__snake_case ) )
# tokenizer config
lowercase_ : Optional[int] = os.path.join(__snake_case , __snake_case )
lowercase_ : str = {
'''bos_token''': '''<s>''',
'''eos_token''': '''</s>''',
'''model_max_length''': 1_0_2_4,
'''pad_token''': '''<pad>''',
'''special_tokens_map_file''': None,
'''tokenizer_class''': '''BioGptTokenizer''',
'''unk_token''': '''<unk>''',
}
print(F'''Generating {biogpt_tokenizer_config_file}''' )
with open(__snake_case , '''w''' , encoding='''utf-8''' ) as f:
f.write(json.dumps(__snake_case , ensure_ascii=__snake_case , indent=__snake_case ) )
# model
lowercase_ : Tuple = chkpt['''model''']
# remove unneeded keys
lowercase_ : Dict = [
'''decoder.version''',
]
for k in ignore_keys:
model_state_dict.pop(__snake_case , __snake_case )
lowercase_ : List[Any] = list(model_state_dict.keys() )
for layer_name in layer_names:
if layer_name.endswith('''output_projection.weight''' ):
lowercase_ : Optional[int] = model_state_dict.pop(__snake_case )
else:
lowercase_ : str = model_state_dict.pop(__snake_case )
lowercase_ : int = BioGptConfig.from_pretrained(__snake_case )
lowercase_ : int = BioGptForCausalLM(__snake_case )
# check that it loads ok
model_new.load_state_dict(__snake_case )
# save
lowercase_ : Optional[int] = os.path.join(__snake_case , __snake_case )
print(F'''Generating {pytorch_weights_dump_path}''' )
torch.save(__snake_case , __snake_case )
print('''Conversion is done!''' )
if __name__ == "__main__":
__A : Any = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'''--biogpt_checkpoint_path''',
default=None,
type=str,
required=True,
help=(
'''Path to the official PyTorch checkpoint file which is expected to reside in the dump dir with dicts,'''
''' bpecodes, etc.'''
),
)
parser.add_argument(
'''--pytorch_dump_folder_path''', default=None, type=str, required=True, help='''Path to the output PyTorch model.'''
)
__A : Dict = parser.parse_args()
convert_biogpt_checkpoint_to_pytorch(args.biogpt_checkpoint_path, args.pytorch_dump_folder_path)
| 33 |
"""simple docstring"""
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __A (snake_case__):
'''simple docstring'''
@slow
@require_torch
def lowerCAmelCase ( self : Union[str, Any] ) ->Dict:
"""simple docstring"""
snake_case_ = EncoderDecoderModel.from_encoder_decoder_pretrained("""prajjwal1/bert-tiny""" , """prajjwal1/bert-tiny""" )
snake_case_ = BertTokenizer.from_pretrained("""bert-base-uncased""" )
snake_case_ = bertabert.config.encoder.vocab_size
snake_case_ = tokenizer.sep_token_id
snake_case_ = tokenizer.cls_token_id
snake_case_ = 128
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""train[:1%]""" )
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""validation[:1%]""" )
snake_case_ = train_dataset.select(range(32 ) )
snake_case_ = val_dataset.select(range(16 ) )
snake_case_ = 4
def _map_to_encoder_decoder_inputs(UpperCAmelCase_ : int ):
# Tokenizer will automatically set [BOS] <text> [EOS]
snake_case_ = tokenizer(batch["""article"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=512 )
snake_case_ = tokenizer(batch["""highlights"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=128 )
snake_case_ = inputs.input_ids
snake_case_ = inputs.attention_mask
snake_case_ = outputs.input_ids
snake_case_ = outputs.input_ids.copy()
snake_case_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch["""labels"""]
]
snake_case_ = outputs.attention_mask
assert all(len(UpperCAmelCase_ ) == 512 for x in inputs.input_ids )
assert all(len(UpperCAmelCase_ ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(UpperCAmelCase_ : Union[str, Any] ):
snake_case_ = pred.label_ids
snake_case_ = pred.predictions
# all unnecessary tokens are removed
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(UpperCAmelCase_ ) )] ) / len(UpperCAmelCase_ )
return {"accuracy": accuracy}
# map train dataset
snake_case_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
train_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
# same for validation dataset
snake_case_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
val_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
snake_case_ = self.get_auto_remove_tmp_dir()
snake_case_ = SeqaSeqTrainingArguments(
output_dir=UpperCAmelCase_ , per_device_train_batch_size=UpperCAmelCase_ , per_device_eval_batch_size=UpperCAmelCase_ , predict_with_generate=UpperCAmelCase_ , evaluation_strategy="""steps""" , do_train=UpperCAmelCase_ , do_eval=UpperCAmelCase_ , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
snake_case_ = SeqaSeqTrainer(
model=UpperCAmelCase_ , args=UpperCAmelCase_ , compute_metrics=_compute_metrics , train_dataset=UpperCAmelCase_ , eval_dataset=UpperCAmelCase_ , tokenizer=UpperCAmelCase_ , )
# start training
trainer.train()
| 347 | 0 |
'''simple docstring'''
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tf_available, is_torch_available
A ={
'configuration_groupvit': [
'GROUPVIT_PRETRAINED_CONFIG_ARCHIVE_MAP',
'GroupViTConfig',
'GroupViTOnnxConfig',
'GroupViTTextConfig',
'GroupViTVisionConfig',
],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
A =[
'GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST',
'GroupViTModel',
'GroupViTPreTrainedModel',
'GroupViTTextModel',
'GroupViTVisionModel',
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
A =[
'TF_GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST',
'TFGroupViTModel',
'TFGroupViTPreTrainedModel',
'TFGroupViTTextModel',
'TFGroupViTVisionModel',
]
if TYPE_CHECKING:
from .configuration_groupvit import (
GROUPVIT_PRETRAINED_CONFIG_ARCHIVE_MAP,
GroupViTConfig,
GroupViTOnnxConfig,
GroupViTTextConfig,
GroupViTVisionConfig,
)
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_groupvit import (
GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST,
GroupViTModel,
GroupViTPreTrainedModel,
GroupViTTextModel,
GroupViTVisionModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_groupvit import (
TF_GROUPVIT_PRETRAINED_MODEL_ARCHIVE_LIST,
TFGroupViTModel,
TFGroupViTPreTrainedModel,
TFGroupViTTextModel,
TFGroupViTVisionModel,
)
else:
import sys
A =_LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 34 |
"""simple docstring"""
# this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.:
# python ./utils/get_modified_files.py utils src tests examples
#
# it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered
# since the output of this script is fed into Makefile commands it doesn't print a newline after the results
import re
import subprocess
import sys
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8')
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split()
__SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:])
__SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""")
__SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)]
print(' '.join(relevant_modified_files), end='')
| 347 | 0 |
'''simple docstring'''
def __snake_case( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase ) -> bool:
return not any(
neighbour == 1 and colored_vertices[i] == color
for i, neighbour in enumerate(_lowerCAmelCase ) )
def __snake_case( _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase ) -> bool:
# Base Case
if index == len(_lowerCAmelCase ):
return True
# Recursive Step
for i in range(_lowerCAmelCase ):
if valid_coloring(graph[index] , _lowerCAmelCase , _lowerCAmelCase ):
# Color current vertex
snake_case__ : List[Any] = i
# Validate coloring
if util_color(_lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , index + 1 ):
return True
# Backtrack
snake_case__ : Optional[Any] = -1
return False
def __snake_case( _lowerCAmelCase , _lowerCAmelCase ) -> list[int]:
snake_case__ : Dict = [-1] * len(_lowerCAmelCase )
if util_color(_lowerCAmelCase , _lowerCAmelCase , _lowerCAmelCase , 0 ):
return colored_vertices
return []
| 35 |
"""simple docstring"""
import argparse
import json
import os
import fairseq
import torch
from fairseq.data import Dictionary
from transformers import (
WavaVecaConfig,
WavaVecaCTCTokenizer,
WavaVecaFeatureExtractor,
WavaVecaForCTC,
WavaVecaForPreTraining,
WavaVecaProcessor,
logging,
)
from transformers.models.wavaveca.modeling_wavaveca import WavaVecaForSequenceClassification
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'post_extract_proj': 'feature_projection.projection',
'encoder.pos_conv.0': 'encoder.pos_conv_embed.conv',
'self_attn.k_proj': 'encoder.layers.*.attention.k_proj',
'self_attn.v_proj': 'encoder.layers.*.attention.v_proj',
'self_attn.q_proj': 'encoder.layers.*.attention.q_proj',
'self_attn.out_proj': 'encoder.layers.*.attention.out_proj',
'self_attn_layer_norm': 'encoder.layers.*.layer_norm',
'fc1': 'encoder.layers.*.feed_forward.intermediate_dense',
'fc2': 'encoder.layers.*.feed_forward.output_dense',
'final_layer_norm': 'encoder.layers.*.final_layer_norm',
'encoder.layer_norm': 'encoder.layer_norm',
'adapter_layer': 'encoder.layers.*.adapter_layer',
'w2v_model.layer_norm': 'feature_projection.layer_norm',
'quantizer.weight_proj': 'quantizer.weight_proj',
'quantizer.vars': 'quantizer.codevectors',
'project_q': 'project_q',
'final_proj': 'project_hid',
'w2v_encoder.proj': 'lm_head',
'mask_emb': 'masked_spec_embed',
'pooling_layer.linear': 'projector',
'pooling_layer.projection': 'classifier',
}
__SCREAMING_SNAKE_CASE : List[Any] = [
'lm_head',
'quantizer.weight_proj',
'quantizer.codevectors',
'project_q',
'project_hid',
'projector',
'classifier',
]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = {}
with open(_SCREAMING_SNAKE_CASE , """r""" ) as file:
for line_number, line in enumerate(_SCREAMING_SNAKE_CASE ):
snake_case_ = line.strip()
if line:
snake_case_ = line.split()
snake_case_ = line_number
snake_case_ = words[0]
snake_case_ = value
return result
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
for attribute in key.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ).shape
elif weight_type is not None and weight_type == "param":
snake_case_ = hf_pointer
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = shape_pointer.shape
# let's reduce dimension
snake_case_ = value[0]
else:
snake_case_ = hf_pointer.shape
if hf_shape != value.shape:
raise ValueError(
f"""Shape of hf {key + "." + weight_type if weight_type is not None else ""} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}""" )
if weight_type == "weight":
snake_case_ = value
elif weight_type == "weight_g":
snake_case_ = value
elif weight_type == "weight_v":
snake_case_ = value
elif weight_type == "bias":
snake_case_ = value
elif weight_type == "param":
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = value
else:
snake_case_ = value
logger.info(f"""{key + "." + weight_type if weight_type is not None else ""} was initialized from {full_name}.""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = """.""".join([key, weight_type] )
elif weight_type is not None and weight_type == "param":
snake_case_ = """.""".join([key, hf_param_name] )
else:
snake_case_ = key
snake_case_ = value if """lm_head""" in full_key else value[0]
__SCREAMING_SNAKE_CASE : int = {
'W_a': 'linear_1.weight',
'W_b': 'linear_2.weight',
'b_a': 'linear_1.bias',
'b_b': 'linear_2.bias',
'ln_W': 'norm.weight',
'ln_b': 'norm.bias',
}
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> List[str]:
snake_case_ = False
for key, mapped_key in MAPPING.items():
snake_case_ = """wav2vec2.""" + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key
if key in name or key.split("""w2v_model.""" )[-1] == name.split(""".""" )[0]:
snake_case_ = True
if "*" in mapped_key:
snake_case_ = name.split(_SCREAMING_SNAKE_CASE )[0].split(""".""" )[-2]
snake_case_ = mapped_key.replace("""*""" , _SCREAMING_SNAKE_CASE )
if "weight_g" in name:
snake_case_ = """weight_g"""
elif "weight_v" in name:
snake_case_ = """weight_v"""
elif "bias" in name:
snake_case_ = """bias"""
elif "weight" in name:
# TODO: don't match quantizer.weight_proj
snake_case_ = """weight"""
else:
snake_case_ = None
if hf_dict is not None:
rename_dict(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
else:
set_recursively(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
return is_used
return is_used
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = []
snake_case_ = fairseq_model.state_dict()
snake_case_ = hf_model.wavaveca.feature_extractor
for name, value in fairseq_dict.items():
snake_case_ = False
if "conv_layers" in name:
load_conv_layer(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , hf_model.config.feat_extract_norm == """group""" , )
snake_case_ = True
else:
snake_case_ = load_wavaveca_layer(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not is_used:
unused_weights.append(_SCREAMING_SNAKE_CASE )
logger.warning(f"""Unused weights: {unused_weights}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
snake_case_ = full_name.split("""conv_layers.""" )[-1]
snake_case_ = name.split(""".""" )
snake_case_ = int(items[0] )
snake_case_ = int(items[1] )
if type_id == 0:
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(_SCREAMING_SNAKE_CASE )
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=False ) -> int:
if config_path is not None:
snake_case_ = WavaVecaConfig.from_pretrained(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaConfig()
if is_seq_class:
snake_case_ = read_txt_into_dict(_SCREAMING_SNAKE_CASE )
snake_case_ = idalabel
snake_case_ = WavaVecaForSequenceClassification(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
feature_extractor.save_pretrained(_SCREAMING_SNAKE_CASE )
elif is_finetuned:
if dict_path:
snake_case_ = Dictionary.load(_SCREAMING_SNAKE_CASE )
# important change bos & pad token id since CTC symbol is <pad> and
# not <s> as in fairseq
snake_case_ = target_dict.pad_index
snake_case_ = target_dict.bos_index
snake_case_ = target_dict.eos_index
snake_case_ = len(target_dict.symbols )
snake_case_ = os.path.join(_SCREAMING_SNAKE_CASE , """vocab.json""" )
if not os.path.isdir(_SCREAMING_SNAKE_CASE ):
logger.error("""--pytorch_dump_folder_path ({}) should be a directory""".format(_SCREAMING_SNAKE_CASE ) )
return
os.makedirs(_SCREAMING_SNAKE_CASE , exist_ok=_SCREAMING_SNAKE_CASE )
snake_case_ = target_dict.indices
# fairseq has the <pad> and <s> switched
snake_case_ = 0
snake_case_ = 1
with open(_SCREAMING_SNAKE_CASE , """w""" , encoding="""utf-8""" ) as vocab_handle:
json.dump(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaCTCTokenizer(
_SCREAMING_SNAKE_CASE , unk_token=target_dict.unk_word , pad_token=target_dict.pad_word , bos_token=target_dict.bos_word , eos_token=target_dict.eos_word , word_delimiter_token="""|""" , do_lower_case=_SCREAMING_SNAKE_CASE , )
snake_case_ = True if config.feat_extract_norm == """layer""" else False
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
snake_case_ = WavaVecaProcessor(feature_extractor=_SCREAMING_SNAKE_CASE , tokenizer=_SCREAMING_SNAKE_CASE )
processor.save_pretrained(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaForCTC(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaForPreTraining(_SCREAMING_SNAKE_CASE )
if is_finetuned or is_seq_class:
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={"""data""": """/""".join(dict_path.split("""/""" )[:-1] )} )
else:
snake_case_ = argparse.Namespace(task="""audio_pretraining""" )
snake_case_ = fairseq.tasks.setup_task(_SCREAMING_SNAKE_CASE )
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] , task=_SCREAMING_SNAKE_CASE )
snake_case_ = model[0].eval()
recursively_load_weights(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , not is_finetuned )
hf_wavavec.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : str = argparse.ArgumentParser()
parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.')
parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to fairseq checkpoint')
parser.add_argument('--dict_path', default=None, type=str, help='Path to dict of fine-tuned model')
parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert')
parser.add_argument(
'--not_finetuned', action='store_true', help='Whether the model to convert is a fine-tuned model or not'
)
parser.add_argument(
'--is_seq_class',
action='store_true',
help='Whether the model to convert is a fine-tuned sequence classification model or not',
)
__SCREAMING_SNAKE_CASE : Any = parser.parse_args()
__SCREAMING_SNAKE_CASE : List[Any] = not args.not_finetuned and not args.is_seq_class
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.config_path,
args.dict_path,
is_finetuned,
args.is_seq_class,
)
| 347 | 0 |
import copy
import fnmatch
import json
import os
import pickle as pkl
import shutil
import sys
import tarfile
import tempfile
from collections import OrderedDict
from contextlib import contextmanager
from functools import partial
from hashlib import shaaaa
from io import BytesIO
from pathlib import Path
from urllib.parse import urlparse
from zipfile import ZipFile, is_zipfile
import cva
import numpy as np
import requests
import wget
from filelock import FileLock
from PIL import Image
from tqdm.auto import tqdm
from yaml import Loader, dump, load
try:
import torch
_snake_case = True
except ImportError:
_snake_case = False
try:
from torch.hub import _get_torch_home
_snake_case = _get_torch_home()
except ImportError:
_snake_case = os.path.expanduser(
os.getenv("TORCH_HOME", os.path.join(os.getenv("XDG_CACHE_HOME", "~/.cache"), "torch"))
)
_snake_case = os.path.join(torch_cache_home, "transformers")
_snake_case = "https://cdn.huggingface.co"
_snake_case = "https://s3.amazonaws.com/models.huggingface.co/bert"
_snake_case = "/".join(str(Path(__file__).resolve()).split("/")[:-1])
_snake_case = os.path.join(PATH, "config.yaml")
_snake_case = os.path.join(PATH, "attributes.txt")
_snake_case = os.path.join(PATH, "objects.txt")
_snake_case = os.getenv("PYTORCH_PRETRAINED_BERT_CACHE", default_cache_path)
_snake_case = os.getenv("PYTORCH_TRANSFORMERS_CACHE", PYTORCH_PRETRAINED_BERT_CACHE)
_snake_case = os.getenv("TRANSFORMERS_CACHE", PYTORCH_TRANSFORMERS_CACHE)
_snake_case = "pytorch_model.bin"
_snake_case = "config.yaml"
def A ( _lowerCamelCase=OBJECTS , _lowerCamelCase=ATTRIBUTES ):
'''simple docstring'''
_lowerCAmelCase : Tuple = []
with open(_lowerCamelCase ) as f:
for object in f.readlines():
vg_classes.append(object.split("," )[0].lower().strip() )
_lowerCAmelCase : Tuple = []
with open(_lowerCamelCase ) as f:
for object in f.readlines():
vg_attrs.append(object.split("," )[0].lower().strip() )
return vg_classes, vg_attrs
def A ( _lowerCamelCase ):
'''simple docstring'''
_lowerCAmelCase : str = OrderedDict()
with open(_lowerCamelCase , "rb" ) as f:
_lowerCAmelCase : Optional[int] = pkl.load(_lowerCamelCase )["model"]
for k in copy.deepcopy(list(ckp.keys() ) ):
_lowerCAmelCase : int = ckp.pop(_lowerCamelCase )
if isinstance(_lowerCamelCase , np.ndarray ):
_lowerCAmelCase : Optional[int] = torch.tensor(_lowerCamelCase )
else:
assert isinstance(_lowerCamelCase , torch.tensor ), type(_lowerCamelCase )
_lowerCAmelCase : int = v
return r
class UpperCAmelCase_ :
lowerCamelCase__ = {}
def __init__( self, __a, __a = "root", __a=0):
'''simple docstring'''
_lowerCAmelCase : str = name
_lowerCAmelCase : List[Any] = level
_lowerCAmelCase : List[str] = {}
for k, v in dictionary.items():
if v is None:
raise ValueError()
_lowerCAmelCase : Union[str, Any] = copy.deepcopy(__a)
_lowerCAmelCase : Tuple = copy.deepcopy(__a)
if isinstance(__a, __a):
_lowerCAmelCase : Any = Config(__a, name=__a, level=level + 1)
_lowerCAmelCase : int = v
setattr(self, __a, __a)
_lowerCAmelCase : List[str] = d
def __repr__( self):
'''simple docstring'''
return str(list((self._pointer.keys())))
def __setattr__( self, __a, __a):
'''simple docstring'''
_lowerCAmelCase : Any = val
_lowerCAmelCase : int = val
_lowerCAmelCase : Tuple = key.split(".")
_lowerCAmelCase : Union[str, Any] = len(__a) - 1
_lowerCAmelCase : Any = self._pointer
if len(__a) > 1:
for i, l in enumerate(__a):
if hasattr(self, __a) and isinstance(getattr(self, __a), __a):
setattr(getattr(self, __a), ".".join(levels[i:]), __a)
if l == last_level:
_lowerCAmelCase : Optional[int] = val
else:
_lowerCAmelCase : str = pointer[l]
def snake_case__ ( self):
'''simple docstring'''
return self._pointer
def snake_case__ ( self, __a, __a):
'''simple docstring'''
with open(f"{file_name}", "w") as stream:
dump(__a, __a)
def snake_case__ ( self, __a, __a):
'''simple docstring'''
with open(f"{file_name}", "w") as stream:
json.dump(__a, __a)
@staticmethod
def snake_case__ ( __a):
'''simple docstring'''
with open(__a) as stream:
_lowerCAmelCase : Dict = load(__a, Loader=__a)
return data
def __str__( self):
'''simple docstring'''
_lowerCAmelCase : List[str] = " "
if self._name != "root":
_lowerCAmelCase : Dict = f"{t * (self._level-1)}{self._name}:\n"
else:
_lowerCAmelCase : str = ""
_lowerCAmelCase : str = self._level
for i, (k, v) in enumerate(self._pointer.items()):
if isinstance(__a, __a):
r += f"{t * (self._level)}{v}\n"
self._level += 1
else:
r += f"{t * (self._level)}{k}: {v} ({type(__a).__name__})\n"
_lowerCAmelCase : Optional[int] = level
return r[:-1]
@classmethod
def snake_case__ ( cls, __a, **__a):
'''simple docstring'''
_lowerCAmelCase , _lowerCAmelCase : Tuple = cls.get_config_dict(__a, **__a)
return cls(__a)
@classmethod
def snake_case__ ( cls, __a, **__a):
'''simple docstring'''
_lowerCAmelCase : Optional[int] = kwargs.pop("cache_dir", __a)
_lowerCAmelCase : str = kwargs.pop("force_download", __a)
_lowerCAmelCase : Optional[Any] = kwargs.pop("resume_download", __a)
_lowerCAmelCase : str = kwargs.pop("proxies", __a)
_lowerCAmelCase : int = kwargs.pop("local_files_only", __a)
if os.path.isdir(__a):
_lowerCAmelCase : str = os.path.join(__a, __a)
elif os.path.isfile(__a) or is_remote_url(__a):
_lowerCAmelCase : List[str] = pretrained_model_name_or_path
else:
_lowerCAmelCase : int = hf_bucket_url(__a, filename=__a, use_cdn=__a)
try:
# Load from URL or cache if already cached
_lowerCAmelCase : Optional[Any] = cached_path(
__a, cache_dir=__a, force_download=__a, proxies=__a, resume_download=__a, local_files_only=__a, )
# Load config dict
if resolved_config_file is None:
raise EnvironmentError
_lowerCAmelCase : Tuple = Config.load_yaml(__a)
except EnvironmentError:
_lowerCAmelCase : List[Any] = "Can't load config for"
raise EnvironmentError(__a)
if resolved_config_file == config_file:
print("loading configuration file from path")
else:
print("loading configuration file cache")
return Config.load_yaml(__a), kwargs
def A ( _lowerCamelCase ):
'''simple docstring'''
_lowerCAmelCase : Dict = torch.load("dump.pt" , map_location=in_tensor.device )
_lowerCAmelCase : List[str] = in_tensor.numpy()
_lowerCAmelCase : Optional[Any] = out_tensor.numpy()[0]
print(na.shape , na[0, 0, :5] )
print(na.shape , na[0, 0, :5] )
assert np.allclose(_lowerCamelCase , _lowerCamelCase , rtol=0.01 , atol=0.1 ), (
F"{sum([1 for x in np.isclose(_lowerCamelCase , _lowerCamelCase , rtol=0.01 , atol=0.1 ).flatten() if x is False] )/len(na.flatten() )*100:.4f} %"
" element-wise mismatch"
)
raise Exception("tensors are all good" )
# Hugging face functions below
def A ( _lowerCamelCase ):
'''simple docstring'''
_lowerCAmelCase : List[Any] = urlparse(_lowerCamelCase )
return parsed.scheme in ("http", "https")
def A ( _lowerCamelCase , _lowerCamelCase , _lowerCamelCase=True ):
'''simple docstring'''
_lowerCAmelCase : str = CLOUDFRONT_DISTRIB_PREFIX if use_cdn else S3_BUCKET_PREFIX
_lowerCAmelCase : Any = "/" not in model_id
if legacy_format:
return F"{endpoint}/{model_id}-{filename}"
else:
return F"{endpoint}/{model_id}/{filename}"
def A ( _lowerCamelCase , _lowerCamelCase , _lowerCamelCase=None , _lowerCamelCase=0 , _lowerCamelCase=None , ):
'''simple docstring'''
_lowerCAmelCase : Optional[int] = "python/{}".format(sys.version.split()[0] )
if _torch_available:
ua += "; torch/{}".format(torch.__version__ )
if isinstance(_lowerCamelCase , _lowerCamelCase ):
ua += "; " + "; ".join("{}/{}".format(_lowerCamelCase , _lowerCamelCase ) for k, v in user_agent.items() )
elif isinstance(_lowerCamelCase , _lowerCamelCase ):
ua += "; " + user_agent
_lowerCAmelCase : int = {"user-agent": ua}
if resume_size > 0:
_lowerCAmelCase : List[Any] = "bytes=%d-" % (resume_size,)
_lowerCAmelCase : Tuple = requests.get(_lowerCamelCase , stream=_lowerCamelCase , proxies=_lowerCamelCase , headers=_lowerCamelCase )
if response.status_code == 416: # Range not satisfiable
return
_lowerCAmelCase : List[str] = response.headers.get("Content-Length" )
_lowerCAmelCase : Any = resume_size + int(_lowerCamelCase ) if content_length is not None else None
_lowerCAmelCase : List[Any] = tqdm(
unit="B" , unit_scale=_lowerCamelCase , total=_lowerCamelCase , initial=_lowerCamelCase , desc="Downloading" , )
for chunk in response.iter_content(chunk_size=1_024 ):
if chunk: # filter out keep-alive new chunks
progress.update(len(_lowerCamelCase ) )
temp_file.write(_lowerCamelCase )
progress.close()
def A ( _lowerCamelCase , _lowerCamelCase=None , _lowerCamelCase=False , _lowerCamelCase=None , _lowerCamelCase=10 , _lowerCamelCase=False , _lowerCamelCase=None , _lowerCamelCase=False , ):
'''simple docstring'''
if cache_dir is None:
_lowerCAmelCase : int = TRANSFORMERS_CACHE
if isinstance(_lowerCamelCase , _lowerCamelCase ):
_lowerCAmelCase : Any = str(_lowerCamelCase )
os.makedirs(_lowerCamelCase , exist_ok=_lowerCamelCase )
_lowerCAmelCase : Union[str, Any] = None
if not local_files_only:
try:
_lowerCAmelCase : List[Any] = requests.head(_lowerCamelCase , allow_redirects=_lowerCamelCase , proxies=_lowerCamelCase , timeout=_lowerCamelCase )
if response.status_code == 200:
_lowerCAmelCase : Any = response.headers.get("ETag" )
except (EnvironmentError, requests.exceptions.Timeout):
# etag is already None
pass
_lowerCAmelCase : int = url_to_filename(_lowerCamelCase , _lowerCamelCase )
# get cache path to put the file
_lowerCAmelCase : Dict = os.path.join(_lowerCamelCase , _lowerCamelCase )
# etag is None = we don't have a connection, or url doesn't exist, or is otherwise inaccessible.
# try to get the last downloaded one
if etag is None:
if os.path.exists(_lowerCamelCase ):
return cache_path
else:
_lowerCAmelCase : int = [
file
for file in fnmatch.filter(os.listdir(_lowerCamelCase ) , filename + ".*" )
if not file.endswith(".json" ) and not file.endswith(".lock" )
]
if len(_lowerCamelCase ) > 0:
return os.path.join(_lowerCamelCase , matching_files[-1] )
else:
# If files cannot be found and local_files_only=True,
# the models might've been found if local_files_only=False
# Notify the user about that
if local_files_only:
raise ValueError(
"Cannot find the requested files in the cached path and outgoing traffic has been"
" disabled. To enable model look-ups and downloads online, set 'local_files_only'"
" to False." )
return None
# From now on, etag is not None.
if os.path.exists(_lowerCamelCase ) and not force_download:
return cache_path
# Prevent parallel downloads of the same file with a lock.
_lowerCAmelCase : int = cache_path + ".lock"
with FileLock(_lowerCamelCase ):
# If the download just completed while the lock was activated.
if os.path.exists(_lowerCamelCase ) and not force_download:
# Even if returning early like here, the lock will be released.
return cache_path
if resume_download:
_lowerCAmelCase : List[str] = cache_path + ".incomplete"
@contextmanager
def _resumable_file_manager():
with open(_lowerCamelCase , "a+b" ) as f:
yield f
_lowerCAmelCase : Optional[int] = _resumable_file_manager
if os.path.exists(_lowerCamelCase ):
_lowerCAmelCase : str = os.stat(_lowerCamelCase ).st_size
else:
_lowerCAmelCase : Union[str, Any] = 0
else:
_lowerCAmelCase : int = partial(tempfile.NamedTemporaryFile , dir=_lowerCamelCase , delete=_lowerCamelCase )
_lowerCAmelCase : List[str] = 0
# Download to temporary file, then copy to cache dir once finished.
# Otherwise you get corrupt cache entries if the download gets interrupted.
with temp_file_manager() as temp_file:
print(
"%s not found in cache or force_download set to True, downloading to %s" , _lowerCamelCase , temp_file.name , )
http_get(
_lowerCamelCase , _lowerCamelCase , proxies=_lowerCamelCase , resume_size=_lowerCamelCase , user_agent=_lowerCamelCase , )
os.replace(temp_file.name , _lowerCamelCase )
_lowerCAmelCase : int = {"url": url, "etag": etag}
_lowerCAmelCase : int = cache_path + ".json"
with open(_lowerCamelCase , "w" ) as meta_file:
json.dump(_lowerCamelCase , _lowerCamelCase )
return cache_path
def A ( _lowerCamelCase , _lowerCamelCase=None ):
'''simple docstring'''
_lowerCAmelCase : List[Any] = url.encode("utf-8" )
_lowerCAmelCase : Union[str, Any] = shaaaa(_lowerCamelCase )
_lowerCAmelCase : str = url_hash.hexdigest()
if etag:
_lowerCAmelCase : Tuple = etag.encode("utf-8" )
_lowerCAmelCase : Optional[Any] = shaaaa(_lowerCamelCase )
filename += "." + etag_hash.hexdigest()
if url.endswith(".h5" ):
filename += ".h5"
return filename
def A ( _lowerCamelCase , _lowerCamelCase=None , _lowerCamelCase=False , _lowerCamelCase=None , _lowerCamelCase=False , _lowerCamelCase=None , _lowerCamelCase=False , _lowerCamelCase=False , _lowerCamelCase=False , ):
'''simple docstring'''
if cache_dir is None:
_lowerCAmelCase : Any = TRANSFORMERS_CACHE
if isinstance(_lowerCamelCase , _lowerCamelCase ):
_lowerCAmelCase : int = str(_lowerCamelCase )
if isinstance(_lowerCamelCase , _lowerCamelCase ):
_lowerCAmelCase : int = str(_lowerCamelCase )
if is_remote_url(_lowerCamelCase ):
# URL, so get it from the cache (downloading if necessary)
_lowerCAmelCase : Tuple = get_from_cache(
_lowerCamelCase , cache_dir=_lowerCamelCase , force_download=_lowerCamelCase , proxies=_lowerCamelCase , resume_download=_lowerCamelCase , user_agent=_lowerCamelCase , local_files_only=_lowerCamelCase , )
elif os.path.exists(_lowerCamelCase ):
# File, and it exists.
_lowerCAmelCase : str = url_or_filename
elif urlparse(_lowerCamelCase ).scheme == "":
# File, but it doesn't exist.
raise EnvironmentError("file {} not found".format(_lowerCamelCase ) )
else:
# Something unknown
raise ValueError("unable to parse {} as a URL or as a local path".format(_lowerCamelCase ) )
if extract_compressed_file:
if not is_zipfile(_lowerCamelCase ) and not tarfile.is_tarfile(_lowerCamelCase ):
return output_path
# Path where we extract compressed archives
# We avoid '.' in dir name and add "-extracted" at the end: "./model.zip" => "./model-zip-extracted/"
_lowerCAmelCase , _lowerCAmelCase : Tuple = os.path.split(_lowerCamelCase )
_lowerCAmelCase : Union[str, Any] = output_file.replace("." , "-" ) + "-extracted"
_lowerCAmelCase : Optional[int] = os.path.join(_lowerCamelCase , _lowerCamelCase )
if os.path.isdir(_lowerCamelCase ) and os.listdir(_lowerCamelCase ) and not force_extract:
return output_path_extracted
# Prevent parallel extractions
_lowerCAmelCase : Any = output_path + ".lock"
with FileLock(_lowerCamelCase ):
shutil.rmtree(_lowerCamelCase , ignore_errors=_lowerCamelCase )
os.makedirs(_lowerCamelCase )
if is_zipfile(_lowerCamelCase ):
with ZipFile(_lowerCamelCase , "r" ) as zip_file:
zip_file.extractall(_lowerCamelCase )
zip_file.close()
elif tarfile.is_tarfile(_lowerCamelCase ):
_lowerCAmelCase : List[Any] = tarfile.open(_lowerCamelCase )
tar_file.extractall(_lowerCamelCase )
tar_file.close()
else:
raise EnvironmentError("Archive format of {} could not be identified".format(_lowerCamelCase ) )
return output_path_extracted
return output_path
def A ( _lowerCamelCase , _lowerCamelCase="," ):
'''simple docstring'''
assert isinstance(_lowerCamelCase , _lowerCamelCase )
if os.path.isfile(_lowerCamelCase ):
with open(_lowerCamelCase ) as f:
_lowerCAmelCase : Optional[int] = eval(f.read() )
else:
_lowerCAmelCase : Union[str, Any] = requests.get(_lowerCamelCase )
try:
_lowerCAmelCase : str = requests.json()
except Exception:
_lowerCAmelCase : str = req.content.decode()
assert data is not None, "could not connect"
try:
_lowerCAmelCase : Optional[Any] = eval(_lowerCamelCase )
except Exception:
_lowerCAmelCase : int = data.split("\n" )
req.close()
return data
def A ( _lowerCamelCase ):
'''simple docstring'''
_lowerCAmelCase : str = requests.get(_lowerCamelCase )
_lowerCAmelCase : Any = np.array(Image.open(BytesIO(response.content ) ) )
return img
def A ( _lowerCamelCase ):
'''simple docstring'''
_lowerCAmelCase : List[str] = url.split("/" )[-1]
if fn not in os.listdir(os.getcwd() ):
wget.download(_lowerCamelCase )
with open(_lowerCamelCase , "rb" ) as stream:
_lowerCAmelCase : Any = pkl.load(_lowerCamelCase )
_lowerCAmelCase : Optional[Any] = weights.pop("model" )
_lowerCAmelCase : Any = {}
for k, v in model.items():
_lowerCAmelCase : Union[str, Any] = torch.from_numpy(_lowerCamelCase )
if "running_var" in k:
_lowerCAmelCase : str = torch.tensor([0] )
_lowerCAmelCase : str = k.replace("running_var" , "num_batches_tracked" )
_lowerCAmelCase : List[str] = zero
return new
def A ( ):
'''simple docstring'''
print(F"{os.path.abspath(os.path.join(_lowerCamelCase , os.pardir ) )}/demo.ipynb" )
def A ( _lowerCamelCase , _lowerCamelCase="RGB" ):
'''simple docstring'''
assert isinstance(_lowerCamelCase , _lowerCamelCase )
if os.path.isfile(_lowerCamelCase ):
_lowerCAmelCase : List[str] = cva.imread(_lowerCamelCase )
else:
_lowerCAmelCase : List[str] = get_image_from_url(_lowerCamelCase )
assert img is not None, F"could not connect to: {im}"
_lowerCAmelCase : Optional[Any] = cva.cvtColor(_lowerCamelCase , cva.COLOR_BGR2RGB )
if input_format == "RGB":
_lowerCAmelCase : Optional[Any] = img[:, :, ::-1]
return img
def A ( _lowerCamelCase , _lowerCamelCase=1 ):
'''simple docstring'''
return (images[i : i + batch] for i in range(0 , len(_lowerCamelCase ) , _lowerCamelCase ))
| 36 |
"""simple docstring"""
import tempfile
import unittest
import numpy as np
import transformers
from transformers import GPTaTokenizer, GPTJConfig, is_flax_available, is_torch_available
from transformers.testing_utils import is_pt_flax_cross_test, require_flax, tooslow
from ...generation.test_flax_utils import FlaxGenerationTesterMixin
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.modeling_flax_pytorch_utils import (
convert_pytorch_state_dict_to_flax,
load_flax_weights_in_pytorch_model,
)
from transformers.models.gptj.modeling_flax_gptj import FlaxGPTJForCausalLM, FlaxGPTJModel
if is_torch_available():
import torch
class __A :
'''simple docstring'''
def __init__( self : Dict , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Any=14 , UpperCAmelCase_ : Union[str, Any]=7 , UpperCAmelCase_ : Tuple=True , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Union[str, Any]=False , UpperCAmelCase_ : Union[str, Any]=True , UpperCAmelCase_ : str=99 , UpperCAmelCase_ : Union[str, Any]=32 , UpperCAmelCase_ : List[Any]=4 , UpperCAmelCase_ : Optional[int]=4 , UpperCAmelCase_ : int=4 , UpperCAmelCase_ : str=37 , UpperCAmelCase_ : Any="gelu" , UpperCAmelCase_ : str=0.1 , UpperCAmelCase_ : Union[str, Any]=0.1 , UpperCAmelCase_ : int=512 , UpperCAmelCase_ : Tuple=0.02 , ) ->List[str]:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_input_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = rotary_dim
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = initializer_range
snake_case_ = None
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
def lowerCAmelCase ( self : int ) ->Optional[int]:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_input_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = GPTJConfig(
vocab_size=self.vocab_size , n_embd=self.hidden_size , n_layer=self.num_hidden_layers , n_head=self.num_attention_heads , n_positions=self.max_position_embeddings , use_cache=UpperCAmelCase_ , bos_token_id=self.bos_token_id , eos_token_id=self.eos_token_id , pad_token_id=self.pad_token_id , rotary_dim=self.rotary_dim , )
return (config, input_ids, input_mask)
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """attention_mask""": attention_mask}
return config, inputs_dict
def lowerCAmelCase ( self : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.ones((input_ids.shape[0], max_decoder_length) , dtype="""i4""" )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , attention_mask=UpperCAmelCase_ , past_key_values=outputs_cache.past_key_values , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
def lowerCAmelCase ( self : Union[str, Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = jnp.concatenate(
[attention_mask, jnp.zeros((attention_mask.shape[0], max_decoder_length - attention_mask.shape[1]) )] , axis=-1 , )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , past_key_values=outputs_cache.past_key_values , attention_mask=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
@require_flax
class __A (snake_case__ , snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Any = (FlaxGPTJModel, FlaxGPTJForCausalLM) if is_flax_available() else ()
__lowercase: List[str] = (FlaxGPTJForCausalLM,) if is_flax_available() else ()
def lowerCAmelCase ( self : Tuple ) ->List[str]:
"""simple docstring"""
snake_case_ = FlaxGPTJModelTester(self )
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] ) ->Any:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward_with_attn_mask(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
@tooslow
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = GPTaTokenizer.from_pretrained("""gpt2""" , pad_token="""<|endoftext|>""" , padding_side="""left""" )
snake_case_ = tokenizer(["""Hello this is a long string""", """Hey"""] , return_tensors="""np""" , padding=UpperCAmelCase_ , truncation=UpperCAmelCase_ )
snake_case_ = FlaxGPTJForCausalLM.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = False
snake_case_ = model.config.eos_token_id
snake_case_ = jax.jit(model.generate )
snake_case_ = jit_generate(
inputs["""input_ids"""] , attention_mask=inputs["""attention_mask"""] , pad_token_id=tokenizer.pad_token_id ).sequences
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = [
"""Hello this is a long string of text.\n\nI'm trying to get the text of the""",
"""Hey, I'm a little late to the party. I'm going to""",
]
self.assertListEqual(UpperCAmelCase_ , UpperCAmelCase_ )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = convert_pytorch_state_dict_to_flax(pt_model.state_dict() , UpperCAmelCase_ )
snake_case_ = fx_state
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
pt_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = model_class.from_pretrained(UpperCAmelCase_ , from_pt=UpperCAmelCase_ )
snake_case_ = fx_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output_loaded, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output_loaded[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = load_flax_weights_in_pytorch_model(UpperCAmelCase_ , fx_model.params )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
# make sure weights are tied in PyTorch
pt_model.tie_weights()
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
fx_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = pt_model_class.from_pretrained(UpperCAmelCase_ , from_flax=UpperCAmelCase_ )
with torch.no_grad():
snake_case_ = pt_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@tooslow
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
'''simple docstring'''
import collections
import inspect
import unittest
from transformers import SwinvaConfig
from transformers.testing_utils import require_torch, require_vision, slow, torch_device
from transformers.utils import cached_property, is_torch_available, is_vision_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, _config_zero_init, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import SwinvaForImageClassification, SwinvaForMaskedImageModeling, SwinvaModel
from transformers.models.swinva.modeling_swinva import SWINV2_PRETRAINED_MODEL_ARCHIVE_LIST
if is_vision_available():
from PIL import Image
from transformers import AutoImageProcessor
class lowerCAmelCase_:
'''simple docstring'''
def __init__( self ,__UpperCAmelCase ,__UpperCAmelCase=13 ,__UpperCAmelCase=32 ,__UpperCAmelCase=2 ,__UpperCAmelCase=3 ,__UpperCAmelCase=16 ,__UpperCAmelCase=[1, 2, 1] ,__UpperCAmelCase=[2, 2, 4] ,__UpperCAmelCase=2 ,__UpperCAmelCase=2.0 ,__UpperCAmelCase=True ,__UpperCAmelCase=0.0 ,__UpperCAmelCase=0.0 ,__UpperCAmelCase=0.1 ,__UpperCAmelCase="gelu" ,__UpperCAmelCase=False ,__UpperCAmelCase=True ,__UpperCAmelCase=0.0_2 ,__UpperCAmelCase=1E-5 ,__UpperCAmelCase=True ,__UpperCAmelCase=None ,__UpperCAmelCase=True ,__UpperCAmelCase=10 ,__UpperCAmelCase=8 ,) -> List[Any]:
lowerCAmelCase__ : Any = parent
lowerCAmelCase__ : Any = batch_size
lowerCAmelCase__ : Union[str, Any] = image_size
lowerCAmelCase__ : int = patch_size
lowerCAmelCase__ : Dict = num_channels
lowerCAmelCase__ : Optional[int] = embed_dim
lowerCAmelCase__ : List[str] = depths
lowerCAmelCase__ : Union[str, Any] = num_heads
lowerCAmelCase__ : str = window_size
lowerCAmelCase__ : Union[str, Any] = mlp_ratio
lowerCAmelCase__ : Dict = qkv_bias
lowerCAmelCase__ : Union[str, Any] = hidden_dropout_prob
lowerCAmelCase__ : Optional[Any] = attention_probs_dropout_prob
lowerCAmelCase__ : Tuple = drop_path_rate
lowerCAmelCase__ : Tuple = hidden_act
lowerCAmelCase__ : Dict = use_absolute_embeddings
lowerCAmelCase__ : List[str] = patch_norm
lowerCAmelCase__ : Optional[int] = layer_norm_eps
lowerCAmelCase__ : str = initializer_range
lowerCAmelCase__ : str = is_training
lowerCAmelCase__ : Tuple = scope
lowerCAmelCase__ : Optional[Any] = use_labels
lowerCAmelCase__ : Dict = type_sequence_label_size
lowerCAmelCase__ : Optional[Any] = encoder_stride
def UpperCAmelCase_ ( self ) -> List[Any]:
lowerCAmelCase__ : Dict = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
lowerCAmelCase__ : Dict = None
if self.use_labels:
lowerCAmelCase__ : Tuple = ids_tensor([self.batch_size] ,self.type_sequence_label_size )
lowerCAmelCase__ : List[str] = self.get_config()
return config, pixel_values, labels
def UpperCAmelCase_ ( self ) -> int:
return SwinvaConfig(
image_size=self.image_size ,patch_size=self.patch_size ,num_channels=self.num_channels ,embed_dim=self.embed_dim ,depths=self.depths ,num_heads=self.num_heads ,window_size=self.window_size ,mlp_ratio=self.mlp_ratio ,qkv_bias=self.qkv_bias ,hidden_dropout_prob=self.hidden_dropout_prob ,attention_probs_dropout_prob=self.attention_probs_dropout_prob ,drop_path_rate=self.drop_path_rate ,hidden_act=self.hidden_act ,use_absolute_embeddings=self.use_absolute_embeddings ,path_norm=self.patch_norm ,layer_norm_eps=self.layer_norm_eps ,initializer_range=self.initializer_range ,encoder_stride=self.encoder_stride ,)
def UpperCAmelCase_ ( self ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ) -> List[Any]:
lowerCAmelCase__ : List[str] = SwinvaModel(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
lowerCAmelCase__ : int = model(__UpperCAmelCase )
lowerCAmelCase__ : int = ((config.image_size // config.patch_size) ** 2) // (4 ** (len(config.depths ) - 1))
lowerCAmelCase__ : Tuple = int(config.embed_dim * 2 ** (len(config.depths ) - 1) )
self.parent.assertEqual(result.last_hidden_state.shape ,(self.batch_size, expected_seq_len, expected_dim) )
def UpperCAmelCase_ ( self ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ) -> List[str]:
lowerCAmelCase__ : int = SwinvaForMaskedImageModeling(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
lowerCAmelCase__ : List[str] = model(__UpperCAmelCase )
self.parent.assertEqual(
result.logits.shape ,(self.batch_size, self.num_channels, self.image_size, self.image_size) )
# test greyscale images
lowerCAmelCase__ : Optional[Any] = 1
lowerCAmelCase__ : int = SwinvaForMaskedImageModeling(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
lowerCAmelCase__ : Optional[Any] = floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
lowerCAmelCase__ : List[str] = model(__UpperCAmelCase )
self.parent.assertEqual(result.logits.shape ,(self.batch_size, 1, self.image_size, self.image_size) )
def UpperCAmelCase_ ( self ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ) -> Union[str, Any]:
lowerCAmelCase__ : List[str] = self.type_sequence_label_size
lowerCAmelCase__ : Dict = SwinvaForImageClassification(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
lowerCAmelCase__ : Tuple = model(__UpperCAmelCase ,labels=__UpperCAmelCase )
self.parent.assertEqual(result.logits.shape ,(self.batch_size, self.type_sequence_label_size) )
def UpperCAmelCase_ ( self ) -> Optional[int]:
lowerCAmelCase__ : Union[str, Any] = self.prepare_config_and_inputs()
lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ : Optional[Any] = config_and_inputs
lowerCAmelCase__ : Any = {"""pixel_values""": pixel_values}
return config, inputs_dict
@require_torch
class lowerCAmelCase_( SCREAMING_SNAKE_CASE_ , SCREAMING_SNAKE_CASE_ , unittest.TestCase ):
'''simple docstring'''
__lowercase : int = (
(SwinvaModel, SwinvaForImageClassification, SwinvaForMaskedImageModeling) if is_torch_available() else ()
)
__lowercase : List[str] = (
{'''feature-extraction''': SwinvaModel, '''image-classification''': SwinvaForImageClassification}
if is_torch_available()
else {}
)
__lowercase : str = False
__lowercase : List[str] = False
__lowercase : Any = False
__lowercase : Dict = False
def UpperCAmelCase_ ( self ) -> Optional[int]:
lowerCAmelCase__ : Dict = SwinvaModelTester(self )
lowerCAmelCase__ : List[str] = ConfigTester(self ,config_class=__UpperCAmelCase ,embed_dim=37 )
def UpperCAmelCase_ ( self ) -> Union[str, Any]:
self.config_tester.create_and_test_config_to_json_string()
self.config_tester.create_and_test_config_to_json_file()
self.config_tester.create_and_test_config_from_and_save_pretrained()
self.config_tester.create_and_test_config_with_num_labels()
self.config_tester.check_config_can_be_init_without_params()
self.config_tester.check_config_arguments_init()
def UpperCAmelCase_ ( self ) -> str:
lowerCAmelCase__ : int = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*__UpperCAmelCase )
@unittest.skip(reason="""Got `CUDA error: misaligned address` with PyTorch 2.0.0.""" )
def UpperCAmelCase_ ( self ) -> int:
pass
@unittest.skip(reason="""Swinv2 does not use inputs_embeds""" )
def UpperCAmelCase_ ( self ) -> int:
pass
def UpperCAmelCase_ ( self ) -> Any:
lowerCAmelCase__ , lowerCAmelCase__ : int = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
lowerCAmelCase__ : Optional[int] = model_class(__UpperCAmelCase )
self.assertIsInstance(model.get_input_embeddings() ,(nn.Module) )
lowerCAmelCase__ : List[str] = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(__UpperCAmelCase ,nn.Linear ) )
def UpperCAmelCase_ ( self ) -> str:
lowerCAmelCase__ , lowerCAmelCase__ : List[Any] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
lowerCAmelCase__ : Optional[Any] = model_class(__UpperCAmelCase )
lowerCAmelCase__ : Union[str, Any] = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
lowerCAmelCase__ : Optional[Any] = [*signature.parameters.keys()]
lowerCAmelCase__ : List[str] = ["""pixel_values"""]
self.assertListEqual(arg_names[:1] ,__UpperCAmelCase )
def UpperCAmelCase_ ( self ) -> int:
lowerCAmelCase__ , lowerCAmelCase__ : Optional[Any] = self.model_tester.prepare_config_and_inputs_for_common()
lowerCAmelCase__ : Any = True
for model_class in self.all_model_classes:
lowerCAmelCase__ : int = True
lowerCAmelCase__ : Any = False
lowerCAmelCase__ : List[Any] = True
lowerCAmelCase__ : Optional[Any] = model_class(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
with torch.no_grad():
lowerCAmelCase__ : Tuple = model(**self._prepare_for_class(__UpperCAmelCase ,__UpperCAmelCase ) )
lowerCAmelCase__ : List[str] = outputs.attentions
lowerCAmelCase__ : List[Any] = len(self.model_tester.depths )
self.assertEqual(len(__UpperCAmelCase ) ,__UpperCAmelCase )
# check that output_attentions also work using config
del inputs_dict["output_attentions"]
lowerCAmelCase__ : int = True
lowerCAmelCase__ : Dict = config.window_size**2
lowerCAmelCase__ : Optional[Any] = model_class(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
with torch.no_grad():
lowerCAmelCase__ : Optional[Any] = model(**self._prepare_for_class(__UpperCAmelCase ,__UpperCAmelCase ) )
lowerCAmelCase__ : List[str] = outputs.attentions
self.assertEqual(len(__UpperCAmelCase ) ,__UpperCAmelCase )
self.assertListEqual(
list(attentions[0].shape[-3:] ) ,[self.model_tester.num_heads[0], window_size_squared, window_size_squared] ,)
lowerCAmelCase__ : List[str] = len(__UpperCAmelCase )
# Check attention is always last and order is fine
lowerCAmelCase__ : Optional[int] = True
lowerCAmelCase__ : Union[str, Any] = True
lowerCAmelCase__ : Tuple = model_class(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
with torch.no_grad():
lowerCAmelCase__ : Any = model(**self._prepare_for_class(__UpperCAmelCase ,__UpperCAmelCase ) )
if hasattr(self.model_tester ,"""num_hidden_states_types""" ):
lowerCAmelCase__ : int = self.model_tester.num_hidden_states_types
else:
# also another +1 for reshaped_hidden_states
lowerCAmelCase__ : Optional[int] = 2
self.assertEqual(out_len + added_hidden_states ,len(__UpperCAmelCase ) )
lowerCAmelCase__ : Any = outputs.attentions
self.assertEqual(len(__UpperCAmelCase ) ,__UpperCAmelCase )
self.assertListEqual(
list(self_attentions[0].shape[-3:] ) ,[self.model_tester.num_heads[0], window_size_squared, window_size_squared] ,)
def UpperCAmelCase_ ( self ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ) -> Tuple:
lowerCAmelCase__ : Union[str, Any] = model_class(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
with torch.no_grad():
lowerCAmelCase__ : List[Any] = model(**self._prepare_for_class(__UpperCAmelCase ,__UpperCAmelCase ) )
lowerCAmelCase__ : Optional[int] = outputs.hidden_states
lowerCAmelCase__ : str = getattr(
self.model_tester ,"""expected_num_hidden_layers""" ,len(self.model_tester.depths ) + 1 )
self.assertEqual(len(__UpperCAmelCase ) ,__UpperCAmelCase )
# Swinv2 has a different seq_length
lowerCAmelCase__ : int = (
config.patch_size
if isinstance(config.patch_size ,collections.abc.Iterable )
else (config.patch_size, config.patch_size)
)
lowerCAmelCase__ : Tuple = (image_size[1] // patch_size[1]) * (image_size[0] // patch_size[0])
self.assertListEqual(
list(hidden_states[0].shape[-2:] ) ,[num_patches, self.model_tester.embed_dim] ,)
lowerCAmelCase__ : Union[str, Any] = outputs.reshaped_hidden_states
self.assertEqual(len(__UpperCAmelCase ) ,__UpperCAmelCase )
lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ , lowerCAmelCase__ : int = reshaped_hidden_states[0].shape
lowerCAmelCase__ : int = (
reshaped_hidden_states[0].view(__UpperCAmelCase ,__UpperCAmelCase ,height * width ).permute(0 ,2 ,1 )
)
self.assertListEqual(
list(reshaped_hidden_states.shape[-2:] ) ,[num_patches, self.model_tester.embed_dim] ,)
def UpperCAmelCase_ ( self ) -> List[str]:
lowerCAmelCase__ , lowerCAmelCase__ : Optional[int] = self.model_tester.prepare_config_and_inputs_for_common()
lowerCAmelCase__ : Optional[int] = (
self.model_tester.image_size
if isinstance(self.model_tester.image_size ,collections.abc.Iterable )
else (self.model_tester.image_size, self.model_tester.image_size)
)
for model_class in self.all_model_classes:
lowerCAmelCase__ : int = True
self.check_hidden_states_output(__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase )
# check that output_hidden_states also work using config
del inputs_dict["output_hidden_states"]
lowerCAmelCase__ : Tuple = True
self.check_hidden_states_output(__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase )
def UpperCAmelCase_ ( self ) -> Union[str, Any]:
lowerCAmelCase__ , lowerCAmelCase__ : List[str] = self.model_tester.prepare_config_and_inputs_for_common()
lowerCAmelCase__ : int = 3
lowerCAmelCase__ : List[Any] = (
self.model_tester.image_size
if isinstance(self.model_tester.image_size ,collections.abc.Iterable )
else (self.model_tester.image_size, self.model_tester.image_size)
)
lowerCAmelCase__ : int = (
config.patch_size
if isinstance(config.patch_size ,collections.abc.Iterable )
else (config.patch_size, config.patch_size)
)
lowerCAmelCase__ : Union[str, Any] = image_size[0] + patch_size[0] - (image_size[0] % patch_size[0])
lowerCAmelCase__ : Dict = image_size[1] + patch_size[1] - (image_size[1] % patch_size[1])
for model_class in self.all_model_classes:
lowerCAmelCase__ : Union[str, Any] = True
self.check_hidden_states_output(__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ,(padded_height, padded_width) )
# check that output_hidden_states also work using config
del inputs_dict["output_hidden_states"]
lowerCAmelCase__ : int = True
self.check_hidden_states_output(__UpperCAmelCase ,__UpperCAmelCase ,__UpperCAmelCase ,(padded_height, padded_width) )
def UpperCAmelCase_ ( self ) -> int:
lowerCAmelCase__ : List[Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_masked_image_modeling(*__UpperCAmelCase )
def UpperCAmelCase_ ( self ) -> Tuple:
lowerCAmelCase__ : Optional[int] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_image_classification(*__UpperCAmelCase )
@slow
def UpperCAmelCase_ ( self ) -> Any:
for model_name in SWINV2_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
lowerCAmelCase__ : List[Any] = SwinvaModel.from_pretrained(__UpperCAmelCase )
self.assertIsNotNone(__UpperCAmelCase )
def UpperCAmelCase_ ( self ) -> Any:
lowerCAmelCase__ , lowerCAmelCase__ : str = self.model_tester.prepare_config_and_inputs_for_common()
lowerCAmelCase__ : List[str] = _config_zero_init(__UpperCAmelCase )
for model_class in self.all_model_classes:
lowerCAmelCase__ : Tuple = model_class(config=__UpperCAmelCase )
for name, param in model.named_parameters():
if "embeddings" not in name and "logit_scale" not in name and param.requires_grad:
self.assertIn(
((param.data.mean() * 1E9).round() / 1E9).item() ,[0.0, 1.0] ,msg=F"""Parameter {name} of model {model_class} seems not properly initialized""" ,)
@require_vision
@require_torch
class lowerCAmelCase_( unittest.TestCase ):
'''simple docstring'''
@cached_property
def UpperCAmelCase_ ( self ) -> Any:
return (
AutoImageProcessor.from_pretrained("""microsoft/swinv2-tiny-patch4-window8-256""" )
if is_vision_available()
else None
)
@slow
def UpperCAmelCase_ ( self ) -> List[str]:
lowerCAmelCase__ : int = SwinvaForImageClassification.from_pretrained("""microsoft/swinv2-tiny-patch4-window8-256""" ).to(
__UpperCAmelCase )
lowerCAmelCase__ : Tuple = self.default_image_processor
lowerCAmelCase__ : int = Image.open("""./tests/fixtures/tests_samples/COCO/000000039769.png""" )
lowerCAmelCase__ : Tuple = image_processor(images=__UpperCAmelCase ,return_tensors="""pt""" ).to(__UpperCAmelCase )
# forward pass
with torch.no_grad():
lowerCAmelCase__ : Tuple = model(**__UpperCAmelCase )
# verify the logits
lowerCAmelCase__ : str = torch.Size((1, 1000) )
self.assertEqual(outputs.logits.shape ,__UpperCAmelCase )
lowerCAmelCase__ : List[str] = torch.tensor([-0.3_9_4_7, -0.4_3_0_6, 0.0_0_2_6] ).to(__UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.logits[0, :3] ,__UpperCAmelCase ,atol=1E-4 ) )
| 37 |
"""simple docstring"""
import copy
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ..auto.configuration_auto import CONFIG_MAPPING
__SCREAMING_SNAKE_CASE : Any = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """upernet"""
def __init__( self : str , UpperCAmelCase_ : List[str]=None , UpperCAmelCase_ : str=512 , UpperCAmelCase_ : int=0.02 , UpperCAmelCase_ : Optional[Any]=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Tuple=0.4 , UpperCAmelCase_ : Tuple=384 , UpperCAmelCase_ : Union[str, Any]=256 , UpperCAmelCase_ : str=1 , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : Dict , ) ->Union[str, Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
if backbone_config is None:
logger.info("""`backbone_config` is `None`. Initializing the config with the default `ResNet` backbone.""" )
snake_case_ = CONFIG_MAPPING["""resnet"""](out_features=["""stage1""", """stage2""", """stage3""", """stage4"""] )
elif isinstance(UpperCAmelCase_ , UpperCAmelCase_ ):
snake_case_ = backbone_config.get("""model_type""" )
snake_case_ = CONFIG_MAPPING[backbone_model_type]
snake_case_ = config_class.from_dict(UpperCAmelCase_ )
snake_case_ = backbone_config
snake_case_ = hidden_size
snake_case_ = initializer_range
snake_case_ = pool_scales
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_in_channels
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = loss_ignore_index
def lowerCAmelCase ( self : str ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = copy.deepcopy(self.__dict__ )
snake_case_ = self.backbone_config.to_dict()
snake_case_ = self.__class__.model_type
return output
| 347 | 0 |
def SCREAMING_SNAKE_CASE_ ( __magic_name__ : int ) -> bool:
"""simple docstring"""
UpperCamelCase :Optional[Any] = (1 + 24 * n) ** 0.5
return ((1 + root) / 6) % 1 == 0
def SCREAMING_SNAKE_CASE_ ( __magic_name__ : int = 5000 ) -> int:
"""simple docstring"""
UpperCamelCase :Optional[int] = [(i * (3 * i - 1)) // 2 for i in range(1 , __magic_name__ )]
for i, pentagonal_i in enumerate(__magic_name__ ):
for j in range(__magic_name__ , len(__magic_name__ ) ):
UpperCamelCase :List[str] = pentagonal_nums[j]
UpperCamelCase :Dict = pentagonal_i + pentagonal_j
UpperCamelCase :Optional[Any] = pentagonal_j - pentagonal_i
if is_pentagonal(__magic_name__ ) and is_pentagonal(__magic_name__ ):
return b
return -1
if __name__ == "__main__":
print(F'''{solution() = }''')
| 38 |
"""simple docstring"""
import os
import shutil
import tempfile
import unittest
import numpy as np
from transformers import AutoTokenizer, BarkProcessor
from transformers.testing_utils import require_torch, slow
@require_torch
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = """ylacombe/bark-small"""
snake_case_ = tempfile.mkdtemp()
snake_case_ = """en_speaker_1"""
snake_case_ = """This is a test string"""
snake_case_ = """speaker_embeddings_path.json"""
snake_case_ = """speaker_embeddings"""
def lowerCAmelCase ( self : List[str] , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
return AutoTokenizer.from_pretrained(self.checkpoint , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
shutil.rmtree(self.tmpdirname )
def lowerCAmelCase ( self : List[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
snake_case_ = BarkProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
@slow
def lowerCAmelCase ( self : Dict ) ->int:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
processor.save_pretrained(
self.tmpdirname , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , speaker_embeddings_directory=self.speaker_embeddings_directory , )
snake_case_ = self.get_tokenizer(bos_token="""(BOS)""" , eos_token="""(EOS)""" )
snake_case_ = BarkProcessor.from_pretrained(
self.tmpdirname , self.speaker_embeddings_dict_path , bos_token="""(BOS)""" , eos_token="""(EOS)""" , )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
def lowerCAmelCase ( self : Optional[Any] ) ->Any:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
snake_case_ = 35
snake_case_ = 2
snake_case_ = 8
snake_case_ = {
"""semantic_prompt""": np.ones(UpperCAmelCase_ ),
"""coarse_prompt""": np.ones((nb_codebooks_coarse, seq_len) ),
"""fine_prompt""": np.ones((nb_codebooks_total, seq_len) ),
}
# test providing already loaded voice_preset
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from npz file
snake_case_ = os.path.join(self.tmpdirname , """file.npz""" )
np.savez(UpperCAmelCase_ , **UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from the hub
snake_case_ = processor(text=self.input_string , voice_preset=self.voice_preset )
def lowerCAmelCase ( self : Tuple ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string )
snake_case_ = tokenizer(
self.input_string , padding="""max_length""" , max_length=256 , add_special_tokens=UpperCAmelCase_ , return_attention_mask=UpperCAmelCase_ , return_token_type_ids=UpperCAmelCase_ , )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key].squeeze().tolist() )
| 347 | 0 |
import pytest
import requests
from datasets.utils.file_utils import http_head
from .utils import OfflineSimulationMode, RequestWouldHangIndefinitelyError, offline
@pytest.mark.integration
def __A ( )-> str:
"""simple docstring"""
with offline(OfflineSimulationMode.CONNECTION_TIMES_OUT ):
with pytest.raises(__lowerCAmelCase ):
requests.request('GET' , 'https://huggingface.co' )
with pytest.raises(requests.exceptions.ConnectTimeout ):
requests.request('GET' , 'https://huggingface.co' , timeout=1.0 )
@pytest.mark.integration
def __A ( )-> List[str]:
"""simple docstring"""
with offline(OfflineSimulationMode.CONNECTION_FAILS ):
with pytest.raises(requests.exceptions.ConnectionError ):
requests.request('GET' , 'https://huggingface.co' )
def __A ( )-> Tuple:
"""simple docstring"""
with offline(OfflineSimulationMode.HF_DATASETS_OFFLINE_SET_TO_1 ):
with pytest.raises(__lowerCAmelCase ):
http_head('https://huggingface.co' )
| 39 |
"""simple docstring"""
import argparse
import json
import os
import sys
import tempfile
import unittest
from argparse import Namespace
from dataclasses import dataclass, field
from enum import Enum
from pathlib import Path
from typing import List, Literal, Optional
import yaml
from transformers import HfArgumentParser, TrainingArguments
from transformers.hf_argparser import make_choice_type_function, string_to_bool
# Since Python 3.10, we can use the builtin `|` operator for Union types
# See PEP 604: https://peps.python.org/pep-0604
__SCREAMING_SNAKE_CASE : int = sys.version_info >= (3, 10)
def _a ( _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
return field(default_factory=lambda: default , metadata=_SCREAMING_SNAKE_CASE )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: float
__lowercase: str
__lowercase: bool
@dataclass
class __A :
'''simple docstring'''
__lowercase: int = 42
__lowercase: str = field(default="""toto""" , metadata={"""help""": """help message"""})
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: Optional[bool] = None
class __A (snake_case__):
'''simple docstring'''
__lowercase: str = """titi"""
__lowercase: Any = """toto"""
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """titi"""
__lowercase: Optional[Any] = """toto"""
__lowercase: List[Any] = 42
@dataclass
class __A :
'''simple docstring'''
__lowercase: BasicEnum = "toto"
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
snake_case_ = BasicEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: MixedTypeEnum = "toto"
def lowerCAmelCase ( self : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = MixedTypeEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: Optional[int] = None
__lowercase: Optional[float] = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: Optional[str] = None
__lowercase: Optional[List[str]] = list_field(default=[])
__lowercase: Optional[List[int]] = list_field(default=[])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = list_field(default=[])
__lowercase: List[int] = list_field(default=[1, 2, 3])
__lowercase: List[str] = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
__lowercase: List[float] = list_field(default=[0.1, 0.2, 0.3])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = field()
__lowercase: str = field()
__lowercase: BasicEnum = field()
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
snake_case_ = BasicEnum(self.required_enum )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: "BasicEnum" = field()
__lowercase: "Optional[bool]" = None
__lowercase: "str" = field(default="""toto""" , metadata={"""help""": """help message"""})
__lowercase: "List[str]" = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
if is_python_no_less_than_3_10:
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: bool | None = None
@dataclass
class __A :
'''simple docstring'''
__lowercase: int | None = None
__lowercase: float | None = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: str | None = None
__lowercase: list[str] | None = list_field(default=[])
__lowercase: list[int] | None = list_field(default=[])
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : argparse.ArgumentParser , UpperCAmelCase_ : argparse.ArgumentParser ) ->Optional[int]:
"""simple docstring"""
self.assertEqual(len(a._actions ) , len(b._actions ) )
for x, y in zip(a._actions , b._actions ):
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
# Choices with mixed type have custom function as "type"
# So we need to compare results directly for equality
if xx.get("""choices""" , UpperCAmelCase_ ) and yy.get("""choices""" , UpperCAmelCase_ ):
for expected_choice in yy["choices"] + xx["choices"]:
self.assertEqual(xx["""type"""](UpperCAmelCase_ ) , yy["""type"""](UpperCAmelCase_ ) )
del xx["type"], yy["type"]
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--bar""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--flag""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = ["""--foo""", """1""", """--baz""", """quux""", """--bar""", """0.5"""]
((snake_case_) , ) = parser.parse_args_into_dataclasses(UpperCAmelCase_ , look_for_args_file=UpperCAmelCase_ )
self.assertFalse(example.flag )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=42 , type=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
# A boolean no_* argument always has to come after its "default: True" regular counter-part
# and its default must be set to False
expected.add_argument("""--no_baz""" , action="""store_false""" , default=UpperCAmelCase_ , dest="""baz""" )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
snake_case_ = [WithDefaultBoolExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--no_baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """True""", """--baz""", """True""", """--opt""", """True"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """False""", """--baz""", """False""", """--opt""", """False"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
def lowerCAmelCase ( self : int ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=["""titi""", """toto""", 42] , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args_into_dataclasses([] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.toto )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """titi"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.titi )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """42"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.fourtytwo )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
@dataclass
class __A :
'''simple docstring'''
__lowercase: Literal["titi", "toto", 42] = "toto"
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=("""titi""", """toto""", 42) , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo_int""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--bar_int""" , nargs="""+""" , default=[1, 2, 3] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_float""" , nargs="""+""" , default=[0.1, 0.2, 0.3] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(
UpperCAmelCase_ , Namespace(foo_int=[] , bar_int=[1, 2, 3] , foo_str=["""Hallo""", """Bonjour""", """Hello"""] , foo_float=[0.1, 0.2, 0.3] ) , )
snake_case_ = parser.parse_args("""--foo_int 1 --bar_int 2 3 --foo_str a b c --foo_float 0.1 0.7""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo_int=[1] , bar_int=[2, 3] , foo_str=["""a""", """b""", """c"""] , foo_float=[0.1, 0.7] ) )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--bar""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--baz""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--ces""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--des""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
snake_case_ = [OptionalExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , bar=UpperCAmelCase_ , baz=UpperCAmelCase_ , ces=[] , des=[] ) )
snake_case_ = parser.parse_args("""--foo 12 --bar 3.14 --baz 42 --ces a b c --des 1 2 3""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=12 , bar=3.14 , baz="""42""" , ces=["""a""", """b""", """c"""] , des=[1, 2, 3] ) )
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--required_list""" , nargs="""+""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--required_str""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
snake_case_ = parser.parse_dict(UpperCAmelCase_ )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
"""extra""": 42,
}
self.assertRaises(UpperCAmelCase_ , parser.parse_dict , UpperCAmelCase_ , allow_extra_keys=UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_json""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.json""" , """w+""" ) as f:
json.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.json""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_yaml""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.yaml""" , """w+""" ) as f:
yaml.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.yaml""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
"""simple docstring"""
# Copyright 2023 The HuggingFace Inc. team. All rights reserved.
#
# 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.
import re
from ..models.auto import AutoProcessor
from ..models.vision_encoder_decoder import VisionEncoderDecoderModel
from ..utils import is_vision_available
from .base import PipelineTool
if is_vision_available():
from PIL import Image
class _A ( _a ):
"""simple docstring"""
UpperCAmelCase : str = """naver-clova-ix/donut-base-finetuned-docvqa"""
UpperCAmelCase : Tuple = (
"""This is a tool that answers a question about an document (pdf). It takes an input named `document` which """
"""should be the document containing the information, as well as a `question` that is the question about the """
"""document. It returns a text that contains the answer to the question."""
)
UpperCAmelCase : List[str] = """document_qa"""
UpperCAmelCase : str = AutoProcessor
UpperCAmelCase : Optional[int] = VisionEncoderDecoderModel
UpperCAmelCase : int = ["""image""", """text"""]
UpperCAmelCase : int = ["""text"""]
def __init__( self : Tuple , *__UpperCAmelCase : Union[str, Any] , **__UpperCAmelCase : Any):
if not is_vision_available():
raise ValueError("Pillow must be installed to use the DocumentQuestionAnsweringTool.")
super().__init__(*__UpperCAmelCase , **__UpperCAmelCase)
def __snake_case ( self : Tuple , __UpperCAmelCase : "Image" , __UpperCAmelCase : str):
a : Any = "<s_docvqa><s_question>{user_input}</s_question><s_answer>"
a : Union[str, Any] = task_prompt.replace("{user_input}" , __UpperCAmelCase)
a : Optional[Any] = self.pre_processor.tokenizer(
__UpperCAmelCase , add_special_tokens=__UpperCAmelCase , return_tensors="pt").input_ids
a : Any = self.pre_processor(__UpperCAmelCase , return_tensors="pt").pixel_values
return {"decoder_input_ids": decoder_input_ids, "pixel_values": pixel_values}
def __snake_case ( self : int , __UpperCAmelCase : int):
return self.model.generate(
inputs["pixel_values"].to(self.device) , decoder_input_ids=inputs["decoder_input_ids"].to(self.device) , max_length=self.model.decoder.config.max_position_embeddings , early_stopping=__UpperCAmelCase , pad_token_id=self.pre_processor.tokenizer.pad_token_id , eos_token_id=self.pre_processor.tokenizer.eos_token_id , use_cache=__UpperCAmelCase , num_beams=1 , bad_words_ids=[[self.pre_processor.tokenizer.unk_token_id]] , return_dict_in_generate=__UpperCAmelCase , ).sequences
def __snake_case ( self : str , __UpperCAmelCase : List[Any]):
a : Union[str, Any] = self.pre_processor.batch_decode(__UpperCAmelCase)[0]
a : Optional[Any] = sequence.replace(self.pre_processor.tokenizer.eos_token , "")
a : Any = sequence.replace(self.pre_processor.tokenizer.pad_token , "")
a : Optional[Any] = re.sub(r"<.*?>" , "" , __UpperCAmelCase , count=1).strip() # remove first task start token
a : List[str] = self.pre_processor.tokenajson(__UpperCAmelCase)
return sequence["answer"]
| 40 |
"""simple docstring"""
import logging
import os
from typing import Dict, List, Optional, Union
import torch
import torch.nn as nn
from accelerate.utils.imports import (
is_abit_bnb_available,
is_abit_bnb_available,
is_bnb_available,
)
from ..big_modeling import dispatch_model, init_empty_weights
from .dataclasses import BnbQuantizationConfig
from .modeling import (
find_tied_parameters,
get_balanced_memory,
infer_auto_device_map,
load_checkpoint_in_model,
offload_weight,
set_module_tensor_to_device,
)
if is_bnb_available():
import bitsandbytes as bnb
from copy import deepcopy
__SCREAMING_SNAKE_CASE : Any = logging.getLogger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = False , ) -> Optional[Any]:
snake_case_ = bnb_quantization_config.load_in_abit
snake_case_ = bnb_quantization_config.load_in_abit
if load_in_abit and not is_abit_bnb_available():
raise ImportError(
"""You have a version of `bitsandbytes` that is not compatible with 8bit quantization,"""
""" make sure you have the latest version of `bitsandbytes` installed.""" )
if load_in_abit and not is_abit_bnb_available():
raise ValueError(
"""You have a version of `bitsandbytes` that is not compatible with 4bit quantization,"""
"""make sure you have the latest version of `bitsandbytes` installed.""" )
snake_case_ = []
# custom device map
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) and len(device_map.keys() ) > 1:
snake_case_ = [key for key, value in device_map.items() if value in ["""disk""", """cpu"""]]
# We keep some modules such as the lm_head in their original dtype for numerical stability reasons
if bnb_quantization_config.skip_modules is None:
snake_case_ = get_keys_to_not_convert(_SCREAMING_SNAKE_CASE )
# add cpu modules to skip modules only for 4-bit modules
if load_in_abit:
bnb_quantization_config.skip_modules.extend(_SCREAMING_SNAKE_CASE )
snake_case_ = bnb_quantization_config.skip_modules
# We add the modules we want to keep in full precision
if bnb_quantization_config.keep_in_fpaa_modules is None:
snake_case_ = []
snake_case_ = bnb_quantization_config.keep_in_fpaa_modules
modules_to_not_convert.extend(_SCREAMING_SNAKE_CASE )
# compatibility with peft
snake_case_ = load_in_abit
snake_case_ = load_in_abit
snake_case_ = get_parameter_device(_SCREAMING_SNAKE_CASE )
if model_device.type != "meta":
# quantization of an already loaded model
logger.warning(
"""It is not recommended to quantize a loaded model. """
"""The model should be instantiated under the `init_empty_weights` context manager.""" )
snake_case_ = replace_with_bnb_layers(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
# convert param to the right dtype
snake_case_ = bnb_quantization_config.torch_dtype
for name, param in model.state_dict().items():
if any(module_to_keep_in_fpaa in name for module_to_keep_in_fpaa in keep_in_fpaa_modules ):
param.to(torch.floataa )
if param.dtype != torch.floataa:
snake_case_ = name.replace(""".weight""" , """""" ).replace(""".bias""" , """""" )
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if param is not None:
param.to(torch.floataa )
elif torch.is_floating_point(_SCREAMING_SNAKE_CASE ):
param.to(_SCREAMING_SNAKE_CASE )
if model_device.type == "cuda":
# move everything to cpu in the first place because we can't do quantization if the weights are already on cuda
model.cuda(torch.cuda.current_device() )
torch.cuda.empty_cache()
elif torch.cuda.is_available():
model.to(torch.cuda.current_device() )
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info(
f"""The model device type is {model_device.type}. However, cuda is needed for quantization."""
"""We move the model to cuda.""" )
return model
elif weights_location is None:
raise RuntimeError(
f"""`weights_location` needs to be the folder path containing the weights of the model, but we found {weights_location} """ )
else:
with init_empty_weights():
snake_case_ = replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
snake_case_ = get_quantized_model_device_map(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , max_memory=_SCREAMING_SNAKE_CASE , no_split_module_classes=_SCREAMING_SNAKE_CASE , )
if offload_state_dict is None and device_map is not None and "disk" in device_map.values():
snake_case_ = True
snake_case_ = any(x in list(device_map.values() ) for x in ["""cpu""", """disk"""] )
load_checkpoint_in_model(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , dtype=bnb_quantization_config.torch_dtype , offload_folder=_SCREAMING_SNAKE_CASE , offload_state_dict=_SCREAMING_SNAKE_CASE , keep_in_fpaa_modules=bnb_quantization_config.keep_in_fpaa_modules , offload_abit_bnb=load_in_abit and offload , )
return dispatch_model(_SCREAMING_SNAKE_CASE , device_map=_SCREAMING_SNAKE_CASE , offload_dir=_SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if device_map is None:
if torch.cuda.is_available():
snake_case_ = {"""""": torch.cuda.current_device()}
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info("""The device_map was not initialized.""" """Setting device_map to `{'':torch.cuda.current_device()}`.""" )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
if device_map not in ["auto", "balanced", "balanced_low_0", "sequential"]:
raise ValueError(
"""If passing a string for `device_map`, please choose 'auto', 'balanced', 'balanced_low_0' or """
"""'sequential'.""" )
snake_case_ = {}
special_dtypes.update(
{
name: bnb_quantization_config.torch_dtype
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.skip_modules )
} )
special_dtypes.update(
{
name: torch.floataa
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.keep_in_fpaa_modules )
} )
snake_case_ = {}
snake_case_ = special_dtypes
snake_case_ = no_split_module_classes
snake_case_ = bnb_quantization_config.target_dtype
# get max_memory for each device.
if device_map != "sequential":
snake_case_ = get_balanced_memory(
_SCREAMING_SNAKE_CASE , low_zero=(device_map == """balanced_low_0""") , max_memory=_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE , )
snake_case_ = max_memory
snake_case_ = infer_auto_device_map(_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
# check if don't have any quantized module on the cpu
snake_case_ = bnb_quantization_config.skip_modules + bnb_quantization_config.keep_in_fpaa_modules
snake_case_ = {
key: device_map[key] for key in device_map.keys() if key not in modules_not_to_convert
}
for device in ["cpu", "disk"]:
if device in device_map_without_some_modules.values():
if bnb_quantization_config.load_in_abit:
raise ValueError(
"""
Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit
the quantized model. If you want to dispatch the model on the CPU or the disk while keeping
these modules in `torch_dtype`, you need to pass a custom `device_map` to
`load_and_quantize_model`. Check
https://huggingface.co/docs/accelerate/main/en/usage_guides/quantization#offload-modules-to-cpu-and-disk
for more details.
""" )
else:
logger.info(
"""Some modules are are offloaded to the CPU or the disk. Note that these modules will be converted to 8-bit""" )
del device_map_without_some_modules
return device_map
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if modules_to_not_convert is None:
snake_case_ = []
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not has_been_replaced:
logger.warning(
"""You are loading your model in 8bit or 4bit but no linear modules were found in your model."""
""" this can happen for some architectures such as gpt2 that uses Conv1D instead of Linear layers."""
""" Please double check your model architecture, or submit an issue on github if you think this is"""
""" a bug.""" )
return model
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , ) -> List[Any]:
snake_case_ = False
for name, module in model.named_children():
if current_key_name is None:
snake_case_ = []
current_key_name.append(_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , nn.Linear ) and name not in modules_to_not_convert:
# Check if the current key is not in the `modules_to_not_convert`
snake_case_ = """.""".join(_SCREAMING_SNAKE_CASE )
snake_case_ = True
for key in modules_to_not_convert:
if (
(key in current_key_name_str) and (key + "." in current_key_name_str)
) or key == current_key_name_str:
snake_case_ = False
break
if proceed:
# Load bnb module with empty weight and replace ``nn.Linear` module
if bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.LinearabitLt(
module.in_features , module.out_features , module.bias is not None , has_fpaa_weights=_SCREAMING_SNAKE_CASE , threshold=bnb_quantization_config.llm_inta_threshold , )
elif bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.Linearabit(
module.in_features , module.out_features , module.bias is not None , bnb_quantization_config.bnb_abit_compute_dtype , compress_statistics=bnb_quantization_config.bnb_abit_use_double_quant , quant_type=bnb_quantization_config.bnb_abit_quant_type , )
else:
raise ValueError("""load_in_8bit and load_in_4bit can't be both False""" )
snake_case_ = module.weight.data
if module.bias is not None:
snake_case_ = module.bias.data
bnb_module.requires_grad_(_SCREAMING_SNAKE_CASE )
setattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = True
if len(list(module.children() ) ) > 0:
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = has_been_replaced | _has_been_replaced
# Remove the last key for recursion
current_key_name.pop(-1 )
return model, has_been_replaced
def _a ( _SCREAMING_SNAKE_CASE ) -> Any:
# Create a copy of the model
with init_empty_weights():
snake_case_ = deepcopy(_SCREAMING_SNAKE_CASE ) # this has 0 cost since it is done inside `init_empty_weights` context manager`
snake_case_ = find_tied_parameters(_SCREAMING_SNAKE_CASE )
# For compatibility with Accelerate < 0.18
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
snake_case_ = sum(list(tied_params.values() ) , [] ) + list(tied_params.keys() )
else:
snake_case_ = sum(_SCREAMING_SNAKE_CASE , [] )
snake_case_ = len(_SCREAMING_SNAKE_CASE ) > 0
# Check if it is a base model
snake_case_ = False
if hasattr(_SCREAMING_SNAKE_CASE , """base_model_prefix""" ):
snake_case_ = not hasattr(_SCREAMING_SNAKE_CASE , model.base_model_prefix )
# Ignore this for base models (BertModel, GPT2Model, etc.)
if (not has_tied_params) and is_base_model:
return []
# otherwise they have an attached head
snake_case_ = list(model.named_children() )
snake_case_ = [list_modules[-1][0]]
# add last module together with tied weights
snake_case_ = set(_SCREAMING_SNAKE_CASE ) - set(_SCREAMING_SNAKE_CASE )
snake_case_ = list(set(_SCREAMING_SNAKE_CASE ) ) + list(_SCREAMING_SNAKE_CASE )
# remove ".weight" from the keys
snake_case_ = [""".weight""", """.bias"""]
snake_case_ = []
for name in list_untouched:
for name_to_remove in names_to_remove:
if name_to_remove in name:
snake_case_ = name.replace(_SCREAMING_SNAKE_CASE , """""" )
filtered_module_names.append(_SCREAMING_SNAKE_CASE )
return filtered_module_names
def _a ( _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for m in model.modules():
if isinstance(_SCREAMING_SNAKE_CASE , bnb.nn.Linearabit ):
return True
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]:
return next(parameter.parameters() ).device
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
# if it is not quantized, we quantize and offload the quantized weights and the SCB stats
if fpaa_statistics is None:
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , 0 , dtype=_SCREAMING_SNAKE_CASE , value=_SCREAMING_SNAKE_CASE )
snake_case_ = param_name
snake_case_ = model
if "." in tensor_name:
snake_case_ = tensor_name.split(""".""" )
for split in splits[:-1]:
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if new_module is None:
raise ValueError(f"""{module} has no attribute {split}.""" )
snake_case_ = new_module
snake_case_ = splits[-1]
# offload weights
snake_case_ = False
offload_weight(module._parameters[tensor_name] , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
if hasattr(module._parameters[tensor_name] , """SCB""" ):
offload_weight(
module._parameters[tensor_name].SCB , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE , )
else:
offload_weight(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
offload_weight(_SCREAMING_SNAKE_CASE , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , """meta""" , dtype=_SCREAMING_SNAKE_CASE , value=torch.empty(*param.size() ) )
| 347 | 0 |
'''simple docstring'''
def SCREAMING_SNAKE_CASE_ (UpperCamelCase ) -> int:
assert column_title.isupper()
lowerCamelCase__ : Dict = 0
lowerCamelCase__ : str = len(UpperCamelCase ) - 1
lowerCamelCase__ : str = 0
while index >= 0:
lowerCamelCase__ : List[str] = (ord(column_title[index] ) - 64) * pow(26 , UpperCamelCase )
answer += value
power += 1
index -= 1
return answer
if __name__ == "__main__":
from doctest import testmod
testmod()
| 41 |
"""simple docstring"""
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__SCREAMING_SNAKE_CASE : str = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'microsoft/beit-base-patch16-224-pt22k': (
'https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json'
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = """beit"""
def __init__( self : List[str] , UpperCAmelCase_ : List[Any]=8_192 , UpperCAmelCase_ : Dict=768 , UpperCAmelCase_ : int=12 , UpperCAmelCase_ : Tuple=12 , UpperCAmelCase_ : List[Any]=3_072 , UpperCAmelCase_ : Tuple="gelu" , UpperCAmelCase_ : Dict=0.0 , UpperCAmelCase_ : List[str]=0.0 , UpperCAmelCase_ : Any=0.02 , UpperCAmelCase_ : Optional[Any]=1E-12 , UpperCAmelCase_ : int=224 , UpperCAmelCase_ : Tuple=16 , UpperCAmelCase_ : List[str]=3 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[str]=False , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[Any]=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Any=True , UpperCAmelCase_ : Dict=[3, 5, 7, 11] , UpperCAmelCase_ : Tuple=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : List[Any]=0.4 , UpperCAmelCase_ : Optional[Any]=256 , UpperCAmelCase_ : Optional[Any]=1 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : List[str] , ) ->Optional[Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = initializer_range
snake_case_ = layer_norm_eps
snake_case_ = image_size
snake_case_ = patch_size
snake_case_ = num_channels
snake_case_ = use_mask_token
snake_case_ = use_absolute_position_embeddings
snake_case_ = use_relative_position_bias
snake_case_ = use_shared_relative_position_bias
snake_case_ = layer_scale_init_value
snake_case_ = drop_path_rate
snake_case_ = use_mean_pooling
# decode head attributes (semantic segmentation)
snake_case_ = out_indices
snake_case_ = pool_scales
# auxiliary head attributes (semantic segmentation)
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = semantic_loss_ignore_index
class __A (snake_case__):
'''simple docstring'''
__lowercase: List[Any] = version.parse("""1.11""")
@property
def lowerCAmelCase ( self : Dict ) ->Mapping[str, Mapping[int, str]]:
"""simple docstring"""
return OrderedDict(
[
("""pixel_values""", {0: """batch""", 1: """num_channels""", 2: """height""", 3: """width"""}),
] )
@property
def lowerCAmelCase ( self : Any ) ->float:
"""simple docstring"""
return 1E-4
| 347 | 0 |
'''simple docstring'''
import gc
import unittest
import numpy as np
import torch
from diffusers import (
AudioDiffusionPipeline,
AutoencoderKL,
DDIMScheduler,
DDPMScheduler,
DiffusionPipeline,
Mel,
UNetaDConditionModel,
UNetaDModel,
)
from diffusers.utils import slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
enable_full_determinism()
class __UpperCAmelCase ( unittest.TestCase ):
def lowerCamelCase ( self ):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
@property
def lowerCamelCase ( self ):
"""simple docstring"""
torch.manual_seed(0 )
_snake_case = UNetaDModel(
sample_size=(32, 64) , in_channels=1 , out_channels=1 , layers_per_block=2 , block_out_channels=(1_28, 1_28) , down_block_types=('AttnDownBlock2D', 'DownBlock2D') , up_block_types=('UpBlock2D', 'AttnUpBlock2D') , )
return model
@property
def lowerCamelCase ( self ):
"""simple docstring"""
torch.manual_seed(0 )
_snake_case = UNetaDConditionModel(
sample_size=(64, 32) , in_channels=1 , out_channels=1 , layers_per_block=2 , block_out_channels=(1_28, 1_28) , down_block_types=('CrossAttnDownBlock2D', 'DownBlock2D') , up_block_types=('UpBlock2D', 'CrossAttnUpBlock2D') , cross_attention_dim=10 , )
return model
@property
def lowerCamelCase ( self ):
"""simple docstring"""
torch.manual_seed(0 )
_snake_case = AutoencoderKL(
sample_size=(1_28, 64) , in_channels=1 , out_channels=1 , latent_channels=1 , layers_per_block=2 , block_out_channels=(1_28, 1_28) , down_block_types=('DownEncoderBlock2D', 'DownEncoderBlock2D') , up_block_types=('UpDecoderBlock2D', 'UpDecoderBlock2D') , )
_snake_case = UNetaDModel(
sample_size=(64, 32) , in_channels=1 , out_channels=1 , layers_per_block=2 , block_out_channels=(1_28, 1_28) , down_block_types=('AttnDownBlock2D', 'DownBlock2D') , up_block_types=('UpBlock2D', 'AttnUpBlock2D') , )
return vqvae, unet
@slow
def lowerCamelCase ( self ):
"""simple docstring"""
_snake_case = 'cpu' # ensure determinism for the device-dependent torch.Generator
_snake_case = Mel(
x_res=self.dummy_unet.config.sample_size[1] , y_res=self.dummy_unet.config.sample_size[0] , )
_snake_case = DDPMScheduler()
_snake_case = AudioDiffusionPipeline(vqvae=lowerCAmelCase_ , unet=self.dummy_unet , mel=lowerCAmelCase_ , scheduler=lowerCAmelCase_ )
_snake_case = pipe.to(lowerCAmelCase_ )
pipe.set_progress_bar_config(disable=lowerCAmelCase_ )
_snake_case = torch.Generator(device=lowerCAmelCase_ ).manual_seed(42 )
_snake_case = pipe(generator=lowerCAmelCase_ , steps=4 )
_snake_case = output.audios[0]
_snake_case = output.images[0]
_snake_case = torch.Generator(device=lowerCAmelCase_ ).manual_seed(42 )
_snake_case = pipe(generator=lowerCAmelCase_ , steps=4 , return_dict=lowerCAmelCase_ )
_snake_case = output[0][0]
assert audio.shape == (1, (self.dummy_unet.config.sample_size[1] - 1) * mel.hop_length)
assert (
image.height == self.dummy_unet.config.sample_size[0]
and image.width == self.dummy_unet.config.sample_size[1]
)
_snake_case = np.frombuffer(image.tobytes() , dtype='uint8' )[:10]
_snake_case = np.frombuffer(image_from_tuple.tobytes() , dtype='uint8' )[:10]
_snake_case = np.array([69, 2_55, 2_55, 2_55, 0, 0, 77, 1_81, 12, 1_27] )
assert np.abs(image_slice.flatten() - expected_slice ).max() == 0
assert np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() == 0
_snake_case = Mel(
x_res=self.dummy_vqvae_and_unet[0].config.sample_size[1] , y_res=self.dummy_vqvae_and_unet[0].config.sample_size[0] , )
_snake_case = DDIMScheduler()
_snake_case = self.dummy_vqvae_and_unet
_snake_case = AudioDiffusionPipeline(
vqvae=self.dummy_vqvae_and_unet[0] , unet=dummy_vqvae_and_unet[1] , mel=lowerCAmelCase_ , scheduler=lowerCAmelCase_ )
_snake_case = pipe.to(lowerCAmelCase_ )
pipe.set_progress_bar_config(disable=lowerCAmelCase_ )
np.random.seed(0 )
_snake_case = np.random.uniform(-1 , 1 , ((dummy_vqvae_and_unet[0].config.sample_size[1] - 1) * mel.hop_length,) )
_snake_case = torch.Generator(device=lowerCAmelCase_ ).manual_seed(42 )
_snake_case = pipe(raw_audio=lowerCAmelCase_ , generator=lowerCAmelCase_ , start_step=5 , steps=10 )
_snake_case = output.images[0]
assert (
image.height == self.dummy_vqvae_and_unet[0].config.sample_size[0]
and image.width == self.dummy_vqvae_and_unet[0].config.sample_size[1]
)
_snake_case = np.frombuffer(image.tobytes() , dtype='uint8' )[:10]
_snake_case = np.array([1_20, 1_17, 1_10, 1_09, 1_38, 1_67, 1_38, 1_48, 1_32, 1_21] )
assert np.abs(image_slice.flatten() - expected_slice ).max() == 0
_snake_case = self.dummy_unet_condition
_snake_case = AudioDiffusionPipeline(
vqvae=self.dummy_vqvae_and_unet[0] , unet=lowerCAmelCase_ , mel=lowerCAmelCase_ , scheduler=lowerCAmelCase_ )
_snake_case = pipe.to(lowerCAmelCase_ )
pipe.set_progress_bar_config(disable=lowerCAmelCase_ )
np.random.seed(0 )
_snake_case = torch.rand((1, 1, 10) )
_snake_case = pipe(generator=lowerCAmelCase_ , encoding=lowerCAmelCase_ )
_snake_case = output.images[0]
_snake_case = np.frombuffer(image.tobytes() , dtype='uint8' )[:10]
_snake_case = np.array([1_07, 1_03, 1_20, 1_27, 1_42, 1_22, 1_13, 1_22, 97, 1_11] )
assert np.abs(image_slice.flatten() - expected_slice ).max() == 0
@slow
@require_torch_gpu
class __UpperCAmelCase ( unittest.TestCase ):
def lowerCamelCase ( self ):
"""simple docstring"""
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCamelCase ( self ):
"""simple docstring"""
_snake_case = torch_device
_snake_case = DiffusionPipeline.from_pretrained('teticio/audio-diffusion-ddim-256' )
_snake_case = pipe.to(lowerCAmelCase_ )
pipe.set_progress_bar_config(disable=lowerCAmelCase_ )
_snake_case = torch.Generator(device=lowerCAmelCase_ ).manual_seed(42 )
_snake_case = pipe(generator=lowerCAmelCase_ )
_snake_case = output.audios[0]
_snake_case = output.images[0]
assert audio.shape == (1, (pipe.unet.config.sample_size[1] - 1) * pipe.mel.hop_length)
assert image.height == pipe.unet.config.sample_size[0] and image.width == pipe.unet.config.sample_size[1]
_snake_case = np.frombuffer(image.tobytes() , dtype='uint8' )[:10]
_snake_case = np.array([1_51, 1_67, 1_54, 1_44, 1_22, 1_34, 1_21, 1_05, 70, 26] )
assert np.abs(image_slice.flatten() - expected_slice ).max() == 0
| 42 |
"""simple docstring"""
import os
import re
import warnings
from shutil import copyfile
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
if TYPE_CHECKING:
from ...tokenization_utils_base import TextInput
from ...utils import logging
__SCREAMING_SNAKE_CASE : Union[str, Any] = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : List[Any] = {'vocab_file': 'spiece.model'}
__SCREAMING_SNAKE_CASE : int = {
'vocab_file': {
't5-small': 'https://huggingface.co/t5-small/resolve/main/spiece.model',
't5-base': 'https://huggingface.co/t5-base/resolve/main/spiece.model',
't5-large': 'https://huggingface.co/t5-large/resolve/main/spiece.model',
't5-3b': 'https://huggingface.co/t5-3b/resolve/main/spiece.model',
't5-11b': 'https://huggingface.co/t5-11b/resolve/main/spiece.model',
}
}
# TODO(PVP) - this should be removed in Transformers v5
__SCREAMING_SNAKE_CASE : Dict = {
't5-small': 512,
't5-base': 512,
't5-large': 512,
't5-3b': 512,
't5-11b': 512,
}
__SCREAMING_SNAKE_CASE : Optional[int] = '▁'
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = VOCAB_FILES_NAMES
__lowercase: Any = PRETRAINED_VOCAB_FILES_MAP
__lowercase: Any = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__lowercase: List[str] = ["""input_ids""", """attention_mask"""]
def __init__( self : Optional[int] , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Optional[Any]="</s>" , UpperCAmelCase_ : Optional[Any]="<unk>" , UpperCAmelCase_ : Any="<pad>" , UpperCAmelCase_ : Tuple=100 , UpperCAmelCase_ : Optional[Any]=None , UpperCAmelCase_ : Optional[Dict[str, Any]] = None , UpperCAmelCase_ : Optional[int]=True , **UpperCAmelCase_ : Dict , ) ->None:
"""simple docstring"""
if extra_ids > 0 and additional_special_tokens is None:
snake_case_ = [F"""<extra_id_{i}>""" for i in range(UpperCAmelCase_ )]
elif extra_ids > 0 and additional_special_tokens is not None:
# Check that we have the right number of extra_id special tokens
snake_case_ = len(set(filter(lambda UpperCAmelCase_ : bool("""extra_id""" in str(UpperCAmelCase_ ) ) , UpperCAmelCase_ ) ) )
if extra_tokens != extra_ids:
raise ValueError(
F"""Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are"""
""" provided to T5Tokenizer. In this case the additional_special_tokens must include the extra_ids"""
""" tokens""" )
if legacy:
logger.warning_once(
F"""You are using the legacy behaviour of the {self.__class__}. This means that tokens that come after special tokens will not be properly handled. We recommend you to"""
""" read the related pull request available at https://github.com/huggingface/transformers/pull/24565""" )
snake_case_ = legacy
snake_case_ = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
eos_token=UpperCAmelCase_ , unk_token=UpperCAmelCase_ , pad_token=UpperCAmelCase_ , extra_ids=UpperCAmelCase_ , additional_special_tokens=UpperCAmelCase_ , sp_model_kwargs=self.sp_model_kwargs , legacy=UpperCAmelCase_ , **UpperCAmelCase_ , )
snake_case_ = vocab_file
snake_case_ = extra_ids
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(UpperCAmelCase_ )
@staticmethod
def lowerCAmelCase ( UpperCAmelCase_ : str , UpperCAmelCase_ : Dict , UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
if pretrained_model_name_or_path in TaTokenizer.max_model_input_sizes:
snake_case_ = TaTokenizer.max_model_input_sizes[pretrained_model_name_or_path]
if init_max_model_length is not None and init_max_model_length != max_model_length:
return init_max_model_length
elif init_max_model_length is None:
warnings.warn(
"""This tokenizer was incorrectly instantiated with a model max length of"""
F""" {deprecated_max_model_length} which will be corrected in Transformers v5.\nFor now, this"""
""" behavior is kept to avoid breaking backwards compatibility when padding/encoding with"""
""" `truncation is True`.\n- Be aware that you SHOULD NOT rely on"""
F""" {pretrained_model_name_or_path} automatically truncating your input to"""
F""" {deprecated_max_model_length} when padding/encoding.\n- If you want to encode/pad to sequences"""
F""" longer than {deprecated_max_model_length} you can either instantiate this tokenizer with"""
""" `model_max_length` or pass `max_length` when encoding/padding.\n- To avoid this warning, please"""
""" instantiate this tokenizer with `model_max_length` set to your preferred value.""" , UpperCAmelCase_ , )
return max_model_length
@property
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
return self.sp_model.get_piece_size() + self._extra_ids
def lowerCAmelCase ( self : Any ) ->Optional[int]:
"""simple docstring"""
snake_case_ = {self.convert_ids_to_tokens(UpperCAmelCase_ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None , UpperCAmelCase_ : bool = False ) ->List[int]:
"""simple docstring"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCAmelCase_ , token_ids_a=UpperCAmelCase_ , already_has_special_tokens=UpperCAmelCase_ )
# normal case: some special tokens
if token_ids_a is None:
return ([0] * len(UpperCAmelCase_ )) + [1]
return ([0] * len(UpperCAmelCase_ )) + [1] + ([0] * len(UpperCAmelCase_ )) + [1]
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
return list(
set(filter(lambda UpperCAmelCase_ : bool(re.search(R"""<extra_id_\d+>""" , UpperCAmelCase_ ) ) is not None , self.additional_special_tokens ) ) )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
return [self._convert_token_to_id(UpperCAmelCase_ ) for token in self.get_sentinel_tokens()]
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : List[int] ) ->List[int]:
"""simple docstring"""
if len(UpperCAmelCase_ ) > 0 and token_ids[-1] == self.eos_token_id:
warnings.warn(
F"""This sequence already has {self.eos_token}. In future versions this behavior may lead to duplicated"""
""" eos tokens being added.""" )
return token_ids
else:
return token_ids + [self.eos_token_id]
def lowerCAmelCase ( self : str , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = [self.eos_token_id]
if token_ids_a is None:
return len(token_ids_a + eos ) * [0]
return len(token_ids_a + eos + token_ids_a + eos ) * [0]
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
if token_ids_a is None:
return token_ids_a
else:
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
return token_ids_a + token_ids_a
def __getstate__( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = self.__dict__.copy()
snake_case_ = None
return state
def __setstate__( self : Optional[Any] , UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = d
# for backward compatibility
if not hasattr(self , """sp_model_kwargs""" ):
snake_case_ = {}
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def lowerCAmelCase ( self : int , UpperCAmelCase_ : "TextInput" , **UpperCAmelCase_ : Tuple ) ->List[str]:
"""simple docstring"""
if not self.legacy:
snake_case_ = SPIECE_UNDERLINE + text.replace(UpperCAmelCase_ , """ """ )
return super().tokenize(UpperCAmelCase_ , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Any ) ->Tuple:
"""simple docstring"""
if not self.legacy:
snake_case_ = text.startswith(UpperCAmelCase_ )
if is_first:
snake_case_ = text[1:]
snake_case_ = self.sp_model.encode(UpperCAmelCase_ , out_type=UpperCAmelCase_ )
if not self.legacy and not is_first and not text.startswith(""" """ ) and tokens[0].startswith(UpperCAmelCase_ ):
snake_case_ = ([tokens[0][1:]] if len(tokens[0] ) > 1 else []) + tokens[1:]
return tokens
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
if token.startswith("""<extra_id_""" ):
snake_case_ = re.match(R"""<extra_id_(\d+)>""" , UpperCAmelCase_ )
snake_case_ = int(match.group(1 ) )
return self.vocab_size - num - 1
return self.sp_model.piece_to_id(UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : Optional[Any] ) ->List[Any]:
"""simple docstring"""
if index < self.sp_model.get_piece_size():
snake_case_ = self.sp_model.IdToPiece(UpperCAmelCase_ )
else:
snake_case_ = F"""<extra_id_{self.vocab_size - 1 - index}>"""
return token
def lowerCAmelCase ( self : List[Any] , UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = []
snake_case_ = """"""
snake_case_ = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(UpperCAmelCase_ ) + token
snake_case_ = True
snake_case_ = []
else:
current_sub_tokens.append(UpperCAmelCase_ )
snake_case_ = False
out_string += self.sp_model.decode(UpperCAmelCase_ )
return out_string.strip()
def lowerCAmelCase ( self : str , UpperCAmelCase_ : str , UpperCAmelCase_ : Optional[str] = None ) ->Tuple[str]:
"""simple docstring"""
if not os.path.isdir(UpperCAmelCase_ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case_ = os.path.join(
UpperCAmelCase_ , (filename_prefix + """-""" if filename_prefix else """""") + VOCAB_FILES_NAMES["""vocab_file"""] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(UpperCAmelCase_ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , UpperCAmelCase_ )
elif not os.path.isfile(self.vocab_file ):
with open(UpperCAmelCase_ , """wb""" ) as fi:
snake_case_ = self.sp_model.serialized_model_proto()
fi.write(UpperCAmelCase_ )
return (out_vocab_file,)
| 347 | 0 |
def lowerCamelCase ( SCREAMING_SNAKE_CASE ):
'''simple docstring'''
if not isinstance(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ):
raise ValueError('''multiplicative_persistence() only accepts integral values''' )
if num < 0:
raise ValueError('''multiplicative_persistence() does not accept negative values''' )
__UpperCamelCase :int = 0
__UpperCamelCase :List[Any] = str(SCREAMING_SNAKE_CASE )
while len(SCREAMING_SNAKE_CASE ) != 1:
__UpperCamelCase :List[Any] = [int(SCREAMING_SNAKE_CASE ) for i in num_string]
__UpperCamelCase :Tuple = 1
for i in range(0 , len(SCREAMING_SNAKE_CASE ) ):
total *= numbers[i]
__UpperCamelCase :Any = str(SCREAMING_SNAKE_CASE )
steps += 1
return steps
def lowerCamelCase ( SCREAMING_SNAKE_CASE ):
'''simple docstring'''
if not isinstance(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ):
raise ValueError('''additive_persistence() only accepts integral values''' )
if num < 0:
raise ValueError('''additive_persistence() does not accept negative values''' )
__UpperCamelCase :Optional[int] = 0
__UpperCamelCase :Optional[int] = str(SCREAMING_SNAKE_CASE )
while len(SCREAMING_SNAKE_CASE ) != 1:
__UpperCamelCase :Optional[int] = [int(SCREAMING_SNAKE_CASE ) for i in num_string]
__UpperCamelCase :Tuple = 0
for i in range(0 , len(SCREAMING_SNAKE_CASE ) ):
total += numbers[i]
__UpperCamelCase :Any = str(SCREAMING_SNAKE_CASE )
steps += 1
return steps
if __name__ == "__main__":
import doctest
doctest.testmod()
| 43 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE = 1_000_000 ) -> int:
snake_case_ = [i - 1 for i in range(limit + 1 )]
for i in range(2 , limit + 1 ):
if phi[i] == i - 1:
for j in range(2 * i , limit + 1 , _SCREAMING_SNAKE_CASE ):
phi[j] -= phi[j] // i
return sum(phi[2 : limit + 1] )
if __name__ == "__main__":
print(solution())
| 347 | 0 |
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import (
OptionalDependencyNotAvailable,
_LazyModule,
is_flax_available,
is_tf_available,
is_torch_available,
)
_a : Tuple = {
'configuration_wav2vec2': ['WAV_2_VEC_2_PRETRAINED_CONFIG_ARCHIVE_MAP', 'Wav2Vec2Config'],
'feature_extraction_wav2vec2': ['Wav2Vec2FeatureExtractor'],
'processing_wav2vec2': ['Wav2Vec2Processor'],
'tokenization_wav2vec2': ['Wav2Vec2CTCTokenizer', 'Wav2Vec2Tokenizer'],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_a : Any = [
'WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST',
'Wav2Vec2ForAudioFrameClassification',
'Wav2Vec2ForCTC',
'Wav2Vec2ForMaskedLM',
'Wav2Vec2ForPreTraining',
'Wav2Vec2ForSequenceClassification',
'Wav2Vec2ForXVector',
'Wav2Vec2Model',
'Wav2Vec2PreTrainedModel',
]
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_a : int = [
'TF_WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST',
'TFWav2Vec2ForCTC',
'TFWav2Vec2Model',
'TFWav2Vec2PreTrainedModel',
'TFWav2Vec2ForSequenceClassification',
]
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_a : int = [
'FlaxWav2Vec2ForCTC',
'FlaxWav2Vec2ForPreTraining',
'FlaxWav2Vec2Model',
'FlaxWav2Vec2PreTrainedModel',
]
if TYPE_CHECKING:
from .configuration_wavaveca import WAV_2_VEC_2_PRETRAINED_CONFIG_ARCHIVE_MAP, WavaVecaConfig
from .feature_extraction_wavaveca import WavaVecaFeatureExtractor
from .processing_wavaveca import WavaVecaProcessor
from .tokenization_wavaveca import WavaVecaCTCTokenizer, WavaVecaTokenizer
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_wavaveca import (
WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST,
WavaVecaForAudioFrameClassification,
WavaVecaForCTC,
WavaVecaForMaskedLM,
WavaVecaForPreTraining,
WavaVecaForSequenceClassification,
WavaVecaForXVector,
WavaVecaModel,
WavaVecaPreTrainedModel,
)
try:
if not is_tf_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_wavaveca import (
TF_WAV_2_VEC_2_PRETRAINED_MODEL_ARCHIVE_LIST,
TFWavaVecaForCTC,
TFWavaVecaForSequenceClassification,
TFWavaVecaModel,
TFWavaVecaPreTrainedModel,
)
try:
if not is_flax_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_tf_wavaveca import (
FlaxWavaVecaForCTC,
FlaxWavaVecaForPreTraining,
FlaxWavaVecaModel,
FlaxWavaVecaPreTrainedModel,
)
else:
import sys
_a : Any = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 44 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
print(f"""Vertex\tShortest Distance from vertex {src}""" )
for i, d in enumerate(_SCREAMING_SNAKE_CASE ):
print(f"""{i}\t\t{d}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
return True
return False
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> list[float]:
snake_case_ = [float("""inf""" )] * vertex_count
snake_case_ = 0.0
for _ in range(vertex_count - 1 ):
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
snake_case_ = distance[u] + w
snake_case_ = check_negative_cycle(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if negative_cycle_exists:
raise Exception("""Negative cycle found""" )
return distance
if __name__ == "__main__":
import doctest
doctest.testmod()
__SCREAMING_SNAKE_CASE : int = int(input('Enter number of vertices: ').strip())
__SCREAMING_SNAKE_CASE : Dict = int(input('Enter number of edges: ').strip())
__SCREAMING_SNAKE_CASE : list[dict[str, int]] = [{} for _ in range(E)]
for i in range(E):
print('Edge ', i + 1)
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE : List[str] = (
int(x)
for x in input('Enter source, destination, weight: ').strip().split(' ')
)
__SCREAMING_SNAKE_CASE : Union[str, Any] = {'src': src, 'dst': dest, 'weight': weight}
__SCREAMING_SNAKE_CASE : Union[str, Any] = int(input('\nEnter shortest path source:').strip())
__SCREAMING_SNAKE_CASE : str = bellman_ford(graph, V, E, source)
print_distance(shortest_distance, 0)
| 347 | 0 |
"""simple docstring"""
class __lowerCAmelCase : # Public class to implement a graph
'''simple docstring'''
def __init__( self , _a , _a , _a ):
__a = row
__a = col
__a = graph
def __UpperCAmelCase ( self , _a , _a , _a ):
return (
0 <= i < self.ROW
and 0 <= j < self.COL
and not visited[i][j]
and self.graph[i][j]
)
def __UpperCAmelCase ( self , _a , _a , _a ):
# Checking all 8 elements surrounding nth element
__a = [-1, -1, -1, 0, 0, 1, 1, 1] # Coordinate order
__a = [-1, 0, 1, -1, 1, -1, 0, 1]
__a = True # Make those cells visited
for k in range(8 ):
if self.is_safe(i + row_nbr[k] , j + col_nbr[k] , _a ):
self.diffs(i + row_nbr[k] , j + col_nbr[k] , _a )
def __UpperCAmelCase ( self ): # And finally, count all islands.
__a = [[False for j in range(self.COL )] for i in range(self.ROW )]
__a = 0
for i in range(self.ROW ):
for j in range(self.COL ):
if visited[i][j] is False and self.graph[i][j] == 1:
self.diffs(_a , _a , _a )
count += 1
return count
| 45 |
"""simple docstring"""
import argparse
import logging
import os
import re
import tensorflow as tf
from transformers import (
AutoConfig,
AutoTokenizer,
DataCollatorForLanguageModeling,
PushToHubCallback,
TFAutoModelForMaskedLM,
create_optimizer,
)
__SCREAMING_SNAKE_CASE : List[str] = logging.getLogger(__name__)
__SCREAMING_SNAKE_CASE : str = tf.data.AUTOTUNE
def _a ( ) -> List[str]:
snake_case_ = argparse.ArgumentParser(description="""Train a masked language model on TPU.""" )
parser.add_argument(
"""--pretrained_model_config""" , type=_SCREAMING_SNAKE_CASE , default="""roberta-base""" , help="""The model config to use. Note that we don't copy the model's weights, only the config!""" , )
parser.add_argument(
"""--tokenizer""" , type=_SCREAMING_SNAKE_CASE , default="""unigram-tokenizer-wikitext""" , help="""The name of the tokenizer to load. We use the pretrained tokenizer to initialize the model's vocab size.""" , )
parser.add_argument(
"""--per_replica_batch_size""" , type=_SCREAMING_SNAKE_CASE , default=8 , help="""Batch size per TPU core.""" , )
parser.add_argument(
"""--no_tpu""" , action="""store_true""" , help="""If set, run on CPU and don't try to initialize a TPU. Useful for debugging on non-TPU instances.""" , )
parser.add_argument(
"""--tpu_name""" , type=_SCREAMING_SNAKE_CASE , help="""Name of TPU resource to initialize. Should be blank on Colab, and 'local' on TPU VMs.""" , default="""local""" , )
parser.add_argument(
"""--tpu_zone""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud zone that TPU resource is located in. Only used for non-Colab TPU nodes.""" , )
parser.add_argument(
"""--gcp_project""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud project name. Only used for non-Colab TPU nodes.""" )
parser.add_argument(
"""--bfloat16""" , action="""store_true""" , help="""Use mixed-precision bfloat16 for training. This is the recommended lower-precision format for TPU.""" , )
parser.add_argument(
"""--train_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to training dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--shuffle_buffer_size""" , type=_SCREAMING_SNAKE_CASE , default=2**18 , help="""Size of the shuffle buffer (in samples)""" , )
parser.add_argument(
"""--eval_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to evaluation dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--num_epochs""" , type=_SCREAMING_SNAKE_CASE , default=1 , help="""Number of epochs to train for.""" , )
parser.add_argument(
"""--learning_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-4 , help="""Learning rate to use for training.""" , )
parser.add_argument(
"""--weight_decay_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-3 , help="""Weight decay rate to use for training.""" , )
parser.add_argument(
"""--max_length""" , type=_SCREAMING_SNAKE_CASE , default=512 , help="""Maximum length of tokenized sequences. Should match the setting used in prepare_tfrecord_shards.py""" , )
parser.add_argument(
"""--mlm_probability""" , type=_SCREAMING_SNAKE_CASE , default=0.15 , help="""Fraction of tokens to mask during training.""" , )
parser.add_argument("""--output_dir""" , type=_SCREAMING_SNAKE_CASE , required=_SCREAMING_SNAKE_CASE , help="""Path to save model checkpoints to.""" )
parser.add_argument("""--hub_model_id""" , type=_SCREAMING_SNAKE_CASE , help="""Model ID to upload to on the Hugging Face Hub.""" )
snake_case_ = parser.parse_args()
return args
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
try:
if args.tpu_name:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver(
args.tpu_name , zone=args.tpu_zone , project=args.gcp_project )
else:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
raise RuntimeError(
"""Couldn't connect to TPU! Most likely you need to specify --tpu_name, --tpu_zone, or """
"""--gcp_project. When running on a TPU VM, use --tpu_name local.""" )
tf.config.experimental_connect_to_cluster(_SCREAMING_SNAKE_CASE )
tf.tpu.experimental.initialize_tpu_system(_SCREAMING_SNAKE_CASE )
return tpu
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = 0
for file in file_list:
snake_case_ = file.split("""/""" )[-1]
snake_case_ = re.search(r"""-\d+-(\d+)\.tfrecord""" , _SCREAMING_SNAKE_CASE ).group(1 )
snake_case_ = int(_SCREAMING_SNAKE_CASE )
num_samples += sample_count
return num_samples
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None ) -> Union[str, Any]:
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.data.Dataset.from_tensor_slices(_SCREAMING_SNAKE_CASE )
if shuffle:
snake_case_ = dataset.shuffle(len(_SCREAMING_SNAKE_CASE ) )
snake_case_ = tf.data.TFRecordDataset(_SCREAMING_SNAKE_CASE , num_parallel_reads=_SCREAMING_SNAKE_CASE )
# TF can't infer the total sample count because it doesn't read all the records yet, so we assert it here
snake_case_ = dataset.apply(tf.data.experimental.assert_cardinality(_SCREAMING_SNAKE_CASE ) )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
if shuffle:
assert shuffle_buffer_size is not None
snake_case_ = dataset.shuffle(args.shuffle_buffer_size )
snake_case_ = dataset.batch(_SCREAMING_SNAKE_CASE , drop_remainder=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.prefetch(_SCREAMING_SNAKE_CASE )
return dataset
def _a ( _SCREAMING_SNAKE_CASE ) -> List[Any]:
if not args.no_tpu:
snake_case_ = initialize_tpu(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.distribute.TPUStrategy(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = tf.distribute.OneDeviceStrategy(device="""/gpu:0""" )
if args.bfloataa:
tf.keras.mixed_precision.set_global_policy("""mixed_bfloat16""" )
snake_case_ = AutoTokenizer.from_pretrained(args.tokenizer )
snake_case_ = AutoConfig.from_pretrained(args.pretrained_model_config )
snake_case_ = tokenizer.vocab_size
snake_case_ = tf.io.gfile.glob(os.path.join(args.train_dataset , """*.tfrecord""" ) )
if not training_records:
raise ValueError(f"""No .tfrecord files found in {args.train_dataset}.""" )
snake_case_ = tf.io.gfile.glob(os.path.join(args.eval_dataset , """*.tfrecord""" ) )
if not eval_records:
raise ValueError(f"""No .tfrecord files found in {args.eval_dataset}.""" )
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = num_train_samples // (args.per_replica_batch_size * strategy.num_replicas_in_sync)
snake_case_ = steps_per_epoch * args.num_epochs
with strategy.scope():
snake_case_ = TFAutoModelForMaskedLM.from_config(_SCREAMING_SNAKE_CASE )
model(model.dummy_inputs ) # Pass some dummy inputs through the model to ensure all the weights are built
snake_case_ , snake_case_ = create_optimizer(
num_train_steps=_SCREAMING_SNAKE_CASE , num_warmup_steps=total_train_steps // 20 , init_lr=args.learning_rate , weight_decay_rate=args.weight_decay_rate , )
# Transformers models compute the right loss for their task by default when labels are passed, and will
# use this for training unless you specify your own loss function in compile().
model.compile(optimizer=_SCREAMING_SNAKE_CASE , metrics=["""accuracy"""] )
def decode_fn(_SCREAMING_SNAKE_CASE ):
snake_case_ = {
"""input_ids""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
"""attention_mask""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
}
return tf.io.parse_single_example(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# Many of the data collators in Transformers are TF-compilable when return_tensors == "tf", so we can
# use their methods in our data pipeline.
snake_case_ = DataCollatorForLanguageModeling(
tokenizer=_SCREAMING_SNAKE_CASE , mlm_probability=args.mlm_probability , mlm=_SCREAMING_SNAKE_CASE , return_tensors="""tf""" )
def mask_with_collator(_SCREAMING_SNAKE_CASE ):
# TF really needs an isin() function
snake_case_ = (
~tf.cast(batch["""attention_mask"""] , tf.bool )
| (batch["""input_ids"""] == tokenizer.cls_token_id)
| (batch["""input_ids"""] == tokenizer.sep_token_id)
)
snake_case_ , snake_case_ = data_collator.tf_mask_tokens(
batch["""input_ids"""] , vocab_size=len(_SCREAMING_SNAKE_CASE ) , mask_token_id=tokenizer.mask_token_id , special_tokens_mask=_SCREAMING_SNAKE_CASE , )
return batch
snake_case_ = args.per_replica_batch_size * strategy.num_replicas_in_sync
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , shuffle_buffer_size=args.shuffle_buffer_size , )
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , )
snake_case_ = []
if args.hub_model_id:
callbacks.append(
PushToHubCallback(output_dir=args.output_dir , hub_model_id=args.hub_model_id , tokenizer=_SCREAMING_SNAKE_CASE ) )
model.fit(
_SCREAMING_SNAKE_CASE , validation_data=_SCREAMING_SNAKE_CASE , epochs=args.num_epochs , callbacks=_SCREAMING_SNAKE_CASE , )
model.save_pretrained(args.output_dir )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : Union[str, Any] = parse_args()
main(args)
| 347 | 0 |
"""simple docstring"""
from math import pow
def UpperCAmelCase__ ( SCREAMING_SNAKE_CASE : int , SCREAMING_SNAKE_CASE : int , SCREAMING_SNAKE_CASE : int , SCREAMING_SNAKE_CASE : int , SCREAMING_SNAKE_CASE : int , ):
'''simple docstring'''
if current_sum == needed_sum:
# If the sum of the powers is equal to needed_sum, then we have a solution.
solutions_count += 1
return current_sum, solutions_count
lowerCAmelCase = int(pow(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE ) )
if current_sum + i_to_n <= needed_sum:
# If the sum of the powers is less than needed_sum, then continue adding powers.
current_sum += i_to_n
lowerCAmelCase , lowerCAmelCase = backtrack(
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , current_number + 1 , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE )
current_sum -= i_to_n
if i_to_n < needed_sum:
# If the power of i is less than needed_sum, then try with the next power.
lowerCAmelCase , lowerCAmelCase = backtrack(
SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , current_number + 1 , SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE )
return current_sum, solutions_count
def UpperCAmelCase__ ( SCREAMING_SNAKE_CASE : int , SCREAMING_SNAKE_CASE : int ):
'''simple docstring'''
if not (1 <= needed_sum <= 10_00 and 2 <= power <= 10):
raise ValueError(
"""Invalid input\n"""
"""needed_sum must be between 1 and 1000, power between 2 and 10.""" )
return backtrack(SCREAMING_SNAKE_CASE , SCREAMING_SNAKE_CASE , 1 , 0 , 0 )[1] # Return the solutions_count
if __name__ == "__main__":
import doctest
doctest.testmod()
| 46 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> float:
if density <= 0:
raise ValueError("""Impossible fluid density""" )
if bulk_modulus <= 0:
raise ValueError("""Impossible bulk modulus""" )
return (bulk_modulus / density) ** 0.5
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
'''simple docstring'''
import tempfile
import torch
from diffusers import (
DEISMultistepScheduler,
DPMSolverMultistepScheduler,
DPMSolverSinglestepScheduler,
UniPCMultistepScheduler,
)
from .test_schedulers import SchedulerCommonTest
class A__ ( A__ ):
A__ = (DEISMultistepScheduler,)
A__ = (('num_inference_steps', 25),)
def A ( self : Optional[int] , **_a : str ) -> str:
'''simple docstring'''
_SCREAMING_SNAKE_CASE ={
'num_train_timesteps': 1000,
'beta_start': 0.00_01,
'beta_end': 0.02,
'beta_schedule': 'linear',
'solver_order': 2,
}
config.update(**_a )
return config
def A ( self : Union[str, Any] , _a : Optional[Any]=0 , **_a : Any ) -> Tuple:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =dict(self.forward_default_kwargs )
_SCREAMING_SNAKE_CASE =kwargs.pop('num_inference_steps' , _a )
_SCREAMING_SNAKE_CASE =self.dummy_sample
_SCREAMING_SNAKE_CASE =0.1 * sample
_SCREAMING_SNAKE_CASE =[residual + 0.2, residual + 0.15, residual + 0.10]
for scheduler_class in self.scheduler_classes:
_SCREAMING_SNAKE_CASE =self.get_scheduler_config(**_a )
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
scheduler.set_timesteps(_a )
# copy over dummy past residuals
_SCREAMING_SNAKE_CASE =dummy_past_residuals[: scheduler.config.solver_order]
with tempfile.TemporaryDirectory() as tmpdirname:
scheduler.save_config(_a )
_SCREAMING_SNAKE_CASE =scheduler_class.from_pretrained(_a )
new_scheduler.set_timesteps(_a )
# copy over dummy past residuals
_SCREAMING_SNAKE_CASE =dummy_past_residuals[: new_scheduler.config.solver_order]
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE =sample, sample
for t in range(_a , time_step + scheduler.config.solver_order + 1 ):
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a , **_a ).prev_sample
_SCREAMING_SNAKE_CASE =new_scheduler.step(_a , _a , _a , **_a ).prev_sample
assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical"
def A ( self : Union[str, Any] ) -> Any:
'''simple docstring'''
pass
def A ( self : Optional[int] , _a : List[Any]=0 , **_a : Optional[int] ) -> List[str]:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =dict(self.forward_default_kwargs )
_SCREAMING_SNAKE_CASE =kwargs.pop('num_inference_steps' , _a )
_SCREAMING_SNAKE_CASE =self.dummy_sample
_SCREAMING_SNAKE_CASE =0.1 * sample
_SCREAMING_SNAKE_CASE =[residual + 0.2, residual + 0.15, residual + 0.10]
for scheduler_class in self.scheduler_classes:
_SCREAMING_SNAKE_CASE =self.get_scheduler_config()
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
scheduler.set_timesteps(_a )
# copy over dummy past residuals (must be after setting timesteps)
_SCREAMING_SNAKE_CASE =dummy_past_residuals[: scheduler.config.solver_order]
with tempfile.TemporaryDirectory() as tmpdirname:
scheduler.save_config(_a )
_SCREAMING_SNAKE_CASE =scheduler_class.from_pretrained(_a )
# copy over dummy past residuals
new_scheduler.set_timesteps(_a )
# copy over dummy past residual (must be after setting timesteps)
_SCREAMING_SNAKE_CASE =dummy_past_residuals[: new_scheduler.config.solver_order]
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a , **_a ).prev_sample
_SCREAMING_SNAKE_CASE =new_scheduler.step(_a , _a , _a , **_a ).prev_sample
assert torch.sum(torch.abs(output - new_output ) ) < 1e-5, "Scheduler outputs are not identical"
def A ( self : int , _a : Optional[Any]=None , **_a : Any ) -> Optional[int]:
'''simple docstring'''
if scheduler is None:
_SCREAMING_SNAKE_CASE =self.scheduler_classes[0]
_SCREAMING_SNAKE_CASE =self.get_scheduler_config(**_a )
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
_SCREAMING_SNAKE_CASE =self.scheduler_classes[0]
_SCREAMING_SNAKE_CASE =self.get_scheduler_config(**_a )
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
_SCREAMING_SNAKE_CASE =10
_SCREAMING_SNAKE_CASE =self.dummy_model()
_SCREAMING_SNAKE_CASE =self.dummy_sample_deter
scheduler.set_timesteps(_a )
for i, t in enumerate(scheduler.timesteps ):
_SCREAMING_SNAKE_CASE =model(_a , _a )
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a ).prev_sample
return sample
def A ( self : Tuple ) -> Dict:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =dict(self.forward_default_kwargs )
_SCREAMING_SNAKE_CASE =kwargs.pop('num_inference_steps' , _a )
for scheduler_class in self.scheduler_classes:
_SCREAMING_SNAKE_CASE =self.get_scheduler_config()
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
_SCREAMING_SNAKE_CASE =self.dummy_sample
_SCREAMING_SNAKE_CASE =0.1 * sample
if num_inference_steps is not None and hasattr(_a , 'set_timesteps' ):
scheduler.set_timesteps(_a )
elif num_inference_steps is not None and not hasattr(_a , 'set_timesteps' ):
_SCREAMING_SNAKE_CASE =num_inference_steps
# copy over dummy past residuals (must be done after set_timesteps)
_SCREAMING_SNAKE_CASE =[residual + 0.2, residual + 0.15, residual + 0.10]
_SCREAMING_SNAKE_CASE =dummy_past_residuals[: scheduler.config.solver_order]
_SCREAMING_SNAKE_CASE =scheduler.timesteps[5]
_SCREAMING_SNAKE_CASE =scheduler.timesteps[6]
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a , **_a ).prev_sample
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a , **_a ).prev_sample
self.assertEqual(output_a.shape , sample.shape )
self.assertEqual(output_a.shape , output_a.shape )
def A ( self : Dict ) -> str:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =DEISMultistepScheduler(**self.get_scheduler_config() )
_SCREAMING_SNAKE_CASE =self.full_loop(scheduler=_a )
_SCREAMING_SNAKE_CASE =torch.mean(torch.abs(_a ) )
assert abs(result_mean.item() - 0.2_39_16 ) < 1e-3
_SCREAMING_SNAKE_CASE =DPMSolverSinglestepScheduler.from_config(scheduler.config )
_SCREAMING_SNAKE_CASE =DPMSolverMultistepScheduler.from_config(scheduler.config )
_SCREAMING_SNAKE_CASE =UniPCMultistepScheduler.from_config(scheduler.config )
_SCREAMING_SNAKE_CASE =DEISMultistepScheduler.from_config(scheduler.config )
_SCREAMING_SNAKE_CASE =self.full_loop(scheduler=_a )
_SCREAMING_SNAKE_CASE =torch.mean(torch.abs(_a ) )
assert abs(result_mean.item() - 0.2_39_16 ) < 1e-3
def A ( self : Tuple ) -> Union[str, Any]:
'''simple docstring'''
for timesteps in [25, 50, 100, 999, 1000]:
self.check_over_configs(num_train_timesteps=_a )
def A ( self : Optional[int] ) -> Dict:
'''simple docstring'''
self.check_over_configs(thresholding=_a )
for order in [1, 2, 3]:
for solver_type in ["logrho"]:
for threshold in [0.5, 1.0, 2.0]:
for prediction_type in ["epsilon", "sample"]:
self.check_over_configs(
thresholding=_a , prediction_type=_a , sample_max_value=_a , algorithm_type='deis' , solver_order=_a , solver_type=_a , )
def A ( self : int ) -> List[Any]:
'''simple docstring'''
for prediction_type in ["epsilon", "v_prediction"]:
self.check_over_configs(prediction_type=_a )
def A ( self : List[Any] ) -> Tuple:
'''simple docstring'''
for algorithm_type in ["deis"]:
for solver_type in ["logrho"]:
for order in [1, 2, 3]:
for prediction_type in ["epsilon", "sample"]:
self.check_over_configs(
solver_order=_a , solver_type=_a , prediction_type=_a , algorithm_type=_a , )
_SCREAMING_SNAKE_CASE =self.full_loop(
solver_order=_a , solver_type=_a , prediction_type=_a , algorithm_type=_a , )
assert not torch.isnan(_a ).any(), "Samples have nan numbers"
def A ( self : Any ) -> str:
'''simple docstring'''
self.check_over_configs(lower_order_final=_a )
self.check_over_configs(lower_order_final=_a )
def A ( self : List[str] ) -> List[Any]:
'''simple docstring'''
for num_inference_steps in [1, 2, 3, 5, 10, 50, 100, 999, 1000]:
self.check_over_forward(num_inference_steps=_a , time_step=0 )
def A ( self : List[Any] ) -> Dict:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =self.full_loop()
_SCREAMING_SNAKE_CASE =torch.mean(torch.abs(_a ) )
assert abs(result_mean.item() - 0.2_39_16 ) < 1e-3
def A ( self : Optional[Any] ) -> Tuple:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =self.full_loop(prediction_type='v_prediction' )
_SCREAMING_SNAKE_CASE =torch.mean(torch.abs(_a ) )
assert abs(result_mean.item() - 0.0_91 ) < 1e-3
def A ( self : Tuple ) -> Dict:
'''simple docstring'''
_SCREAMING_SNAKE_CASE =self.scheduler_classes[0]
_SCREAMING_SNAKE_CASE =self.get_scheduler_config(thresholding=_a , dynamic_thresholding_ratio=0 )
_SCREAMING_SNAKE_CASE =scheduler_class(**_a )
_SCREAMING_SNAKE_CASE =10
_SCREAMING_SNAKE_CASE =self.dummy_model()
_SCREAMING_SNAKE_CASE =self.dummy_sample_deter.half()
scheduler.set_timesteps(_a )
for i, t in enumerate(scheduler.timesteps ):
_SCREAMING_SNAKE_CASE =model(_a , _a )
_SCREAMING_SNAKE_CASE =scheduler.step(_a , _a , _a ).prev_sample
assert sample.dtype == torch.floataa
| 47 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE ) -> bool:
if num < 0:
return False
snake_case_ = num
snake_case_ = 0
while num > 0:
snake_case_ = rev_num * 10 + (num % 10)
num //= 10
return num_copy == rev_num
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_sentencepiece_available
SCREAMING_SNAKE_CASE__ : Optional[Any] = {}
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
SCREAMING_SNAKE_CASE__ : Optional[int] = ['BartphoTokenizer']
if TYPE_CHECKING:
try:
if not is_sentencepiece_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .tokenization_bartpho import BartphoTokenizer
else:
import sys
SCREAMING_SNAKE_CASE__ : Optional[int] = _LazyModule(__name__, globals()['__file__'], _import_structure, module_spec=__spec__)
| 48 |
"""simple docstring"""
import unittest
from transformers import SPIECE_UNDERLINE
from transformers.models.speechta import SpeechTaTokenizer
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.tokenization_utils import AddedToken
from ...test_tokenization_common import TokenizerTesterMixin
__SCREAMING_SNAKE_CASE : Tuple = get_tests_dir('fixtures/test_sentencepiece_bpe_char.model')
@require_sentencepiece
@require_tokenizers
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Tuple = SpeechTaTokenizer
__lowercase: int = False
__lowercase: List[str] = True
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
snake_case_ = SpeechTaTokenizer(UpperCAmelCase_ )
snake_case_ = AddedToken("""<mask>""" , lstrip=UpperCAmelCase_ , rstrip=UpperCAmelCase_ )
snake_case_ = mask_token
tokenizer.add_special_tokens({"""mask_token""": mask_token} )
tokenizer.add_tokens(["""<ctc_blank>"""] )
tokenizer.save_pretrained(self.tmpdirname )
def lowerCAmelCase ( self : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = """this is a test"""
snake_case_ = """this is a test"""
return input_text, output_text
def lowerCAmelCase ( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : Any=False , UpperCAmelCase_ : Tuple=20 , UpperCAmelCase_ : Dict=5 ) ->List[Any]:
"""simple docstring"""
snake_case_ , snake_case_ = self.get_input_output_texts(UpperCAmelCase_ )
snake_case_ = tokenizer.encode(UpperCAmelCase_ , add_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.decode(UpperCAmelCase_ , clean_up_tokenization_spaces=UpperCAmelCase_ )
return text, ids
def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = """<pad>"""
snake_case_ = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(UpperCAmelCase_ ) , UpperCAmelCase_ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(UpperCAmelCase_ ) , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-4] , """œ""" )
self.assertEqual(vocab_keys[-2] , """<mask>""" )
self.assertEqual(vocab_keys[-1] , """<ctc_blank>""" )
self.assertEqual(len(UpperCAmelCase_ ) , 81 )
def lowerCAmelCase ( self : Optional[int] ) ->int:
"""simple docstring"""
self.assertEqual(self.get_tokenizer().vocab_size , 79 )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = self.get_tokenizers(do_lower_case=UpperCAmelCase_ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
snake_case_ = ["""aaaaa bbbbbb""", """cccccccccdddddddd"""]
snake_case_ = tokenizer.add_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode("""aaaaa bbbbbb low cccccccccdddddddd l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
snake_case_ = {"""eos_token""": """>>>>|||<||<<|<<""", """pad_token""": """<<<<<|||>|>>>>|>"""}
snake_case_ = tokenizer.add_special_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size_a + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode(
""">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->List[str]:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = tokenizer.tokenize("""This is a test""" )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [SPIECE_UNDERLINE, """T""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """a""", SPIECE_UNDERLINE, """t""", """e""", """s""", """t"""] )
# fmt: on
self.assertListEqual(
tokenizer.convert_tokens_to_ids(UpperCAmelCase_ ) , [4, 32, 11, 10, 12, 4, 10, 12, 4, 7, 4, 6, 5, 12, 6] , )
snake_case_ = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """92000""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
snake_case_ = tokenizer.convert_tokens_to_ids(UpperCAmelCase_ )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [4, 30, 4, 20, 7, 12, 4, 25, 8, 13, 9, 4, 10, 9, 4, 3, 23, 4, 7, 9, 14, 4, 6, 11, 10, 12, 4, 10, 12, 4, 19, 7, 15, 12, 73, 26] )
# fmt: on
snake_case_ = tokenizer.convert_ids_to_tokens(UpperCAmelCase_ )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """<unk>""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = [
"""Transformers (formerly known as pytorch-transformers and pytorch-pretrained-bert) provides """
"""general-purpose architectures (BERT, GPT, RoBERTa, XLM, DistilBert, XLNet...) for Natural """
"""Language Understanding (NLU) and Natural Language Generation (NLG) with over thirty-two pretrained """
"""models in one hundred plus languages and deep interoperability between Jax, PyTorch and TensorFlow.""",
"""BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly """
"""conditioning on both left and right context in all layers.""",
"""The quick brown fox jumps over the lazy dog.""",
]
# fmt: off
snake_case_ = {
"""input_ids""": [
[4, 32, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 64, 19, 8, 13, 18, 5, 13, 15, 22, 4, 28, 9, 8, 20, 9, 4, 7, 12, 4, 24, 22, 6, 8, 13, 17, 11, 39, 6, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 7, 9, 14, 4, 24, 22, 6, 8, 13, 17, 11, 39, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 39, 25, 5, 13, 6, 63, 4, 24, 13, 8, 27, 10, 14, 5, 12, 4, 21, 5, 9, 5, 13, 7, 15, 39, 24, 16, 13, 24, 8, 12, 5, 4, 7, 13, 17, 11, 10, 6, 5, 17, 6, 16, 13, 5, 12, 4, 64, 40, 47, 54, 32, 23, 4, 53, 49, 32, 23, 4, 54, 8, 40, 47, 54, 32, 7, 23, 4, 69, 52, 43, 23, 4, 51, 10, 12, 6, 10, 15, 40, 5, 13, 6, 23, 4, 69, 52, 48, 5, 6, 26, 26, 26, 63, 4, 19, 8, 13, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 61, 9, 14, 5, 13, 12, 6, 7, 9, 14, 10, 9, 21, 4, 64, 48, 52, 61, 63, 4, 7, 9, 14, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 53, 5, 9, 5, 13, 7, 6, 10, 8, 9, 4, 64, 48, 52, 53, 63, 4, 20, 10, 6, 11, 4, 8, 27, 5, 13, 4, 6, 11, 10, 13, 6, 22, 39, 6, 20, 8, 4, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 4, 18, 8, 14, 5, 15, 12, 4, 10, 9, 4, 8, 9, 5, 4, 11, 16, 9, 14, 13, 5, 14, 4, 24, 15, 16, 12, 4, 15, 7, 9, 21, 16, 7, 21, 5, 12, 4, 7, 9, 14, 4, 14, 5, 5, 24, 4, 10, 9, 6, 5, 13, 8, 24, 5, 13, 7, 25, 10, 15, 10, 6, 22, 4, 25, 5, 6, 20, 5, 5, 9, 4, 58, 7, 37, 23, 4, 49, 22, 32, 8, 13, 17, 11, 4, 7, 9, 14, 4, 32, 5, 9, 12, 8, 13, 55, 15, 8, 20, 26, 2],
[4, 40, 47, 54, 32, 4, 10, 12, 4, 14, 5, 12, 10, 21, 9, 5, 14, 4, 6, 8, 4, 24, 13, 5, 39, 6, 13, 7, 10, 9, 4, 14, 5, 5, 24, 4, 25, 10, 14, 10, 13, 5, 17, 6, 10, 8, 9, 7, 15, 4, 13, 5, 24, 13, 5, 12, 5, 9, 6, 7, 6, 10, 8, 9, 12, 4, 19, 13, 8, 18, 4, 16, 9, 15, 7, 25, 5, 15, 5, 14, 4, 6, 5, 37, 6, 4, 25, 22, 4, 46, 8, 10, 9, 6, 15, 22, 4, 17, 8, 9, 14, 10, 6, 10, 8, 9, 10, 9, 21, 4, 8, 9, 4, 25, 8, 6, 11, 4, 15, 5, 19, 6, 4, 7, 9, 14, 4, 13, 10, 21, 11, 6, 4, 17, 8, 9, 6, 5, 37, 6, 4, 10, 9, 4, 7, 15, 15, 4, 15, 7, 22, 5, 13, 12, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[4, 32, 11, 5, 4, 45, 16, 10, 17, 28, 4, 25, 13, 8, 20, 9, 4, 19, 8, 37, 4, 46, 16, 18, 24, 12, 4, 8, 27, 5, 13, 4, 6, 11, 5, 4, 15, 7, 57, 22, 4, 14, 8, 21, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
],
"""attention_mask""": [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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],
]
}
# fmt: on
self.tokenizer_integration_test_util(
expected_encoding=UpperCAmelCase_ , model_name="""microsoft/speecht5_asr""" , revision="""c5ef64c71905caeccde0e4462ef3f9077224c524""" , sequences=UpperCAmelCase_ , )
| 347 | 0 |
import unittest
import numpy as np
from transformers.testing_utils import require_torch, require_vision
from transformers.utils import is_torch_available, is_vision_available
from ...test_image_processing_common import ImageProcessingSavingTestMixin, prepare_image_inputs
if is_torch_available():
import torch
if is_vision_available():
from PIL import Image
from transformers import PoolFormerImageProcessor
class _A ( unittest.TestCase ):
def __init__( self : Optional[int] , __SCREAMING_SNAKE_CASE : Tuple , __SCREAMING_SNAKE_CASE : Tuple=7 , __SCREAMING_SNAKE_CASE : Union[str, Any]=3 , __SCREAMING_SNAKE_CASE : str=30 , __SCREAMING_SNAKE_CASE : Union[str, Any]=400 , __SCREAMING_SNAKE_CASE : str=True , __SCREAMING_SNAKE_CASE : Optional[int]=None , __SCREAMING_SNAKE_CASE : List[Any]=0.9 , __SCREAMING_SNAKE_CASE : Tuple=None , __SCREAMING_SNAKE_CASE : List[Any]=True , __SCREAMING_SNAKE_CASE : List[str]=[0.5, 0.5, 0.5] , __SCREAMING_SNAKE_CASE : Dict=[0.5, 0.5, 0.5] , ):
'''simple docstring'''
__a = size if size is not None else {'''shortest_edge''': 30}
__a = crop_size if crop_size is not None else {'''height''': 30, '''width''': 30}
__a = parent
__a = batch_size
__a = num_channels
__a = min_resolution
__a = max_resolution
__a = do_resize_and_center_crop
__a = size
__a = crop_pct
__a = crop_size
__a = do_normalize
__a = image_mean
__a = image_std
def _lowerCamelCase ( self : Any):
'''simple docstring'''
return {
"size": self.size,
"do_resize_and_center_crop": self.do_resize_and_center_crop,
"crop_pct": self.crop_pct,
"crop_size": self.crop_size,
"do_normalize": self.do_normalize,
"image_mean": self.image_mean,
"image_std": self.image_std,
}
@require_torch
@require_vision
class _A ( __UpperCAmelCase ,unittest.TestCase ):
UpperCamelCase__ : Union[str, Any] = PoolFormerImageProcessor if is_vision_available() else None
def _lowerCamelCase ( self : Dict):
'''simple docstring'''
__a = PoolFormerImageProcessingTester(self)
@property
def _lowerCamelCase ( self : List[Any]):
'''simple docstring'''
return self.image_processor_tester.prepare_image_processor_dict()
def _lowerCamelCase ( self : List[Any]):
'''simple docstring'''
__a = self.image_processing_class(**self.image_processor_dict)
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''do_resize_and_center_crop'''))
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''size'''))
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''crop_pct'''))
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''do_normalize'''))
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''image_mean'''))
self.assertTrue(hasattr(__SCREAMING_SNAKE_CASE , '''image_std'''))
def _lowerCamelCase ( self : Optional[int]):
'''simple docstring'''
__a = self.image_processing_class.from_dict(self.image_processor_dict)
self.assertEqual(image_processor.size , {'''shortest_edge''': 30})
self.assertEqual(image_processor.crop_size , {'''height''': 30, '''width''': 30})
__a = self.image_processing_class.from_dict(self.image_processor_dict , size=42 , crop_size=84)
self.assertEqual(image_processor.size , {'''shortest_edge''': 42})
self.assertEqual(image_processor.crop_size , {'''height''': 84, '''width''': 84})
def _lowerCamelCase ( self : List[str]):
'''simple docstring'''
pass
def _lowerCamelCase ( self : Any):
'''simple docstring'''
__a = self.image_processing_class(**self.image_processor_dict)
# create random PIL images
__a = prepare_image_inputs(self.image_processor_tester , equal_resolution=__SCREAMING_SNAKE_CASE)
for image in image_inputs:
self.assertIsInstance(__SCREAMING_SNAKE_CASE , Image.Image)
# Test not batched input
__a = image_processing(image_inputs[0] , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
__a = image_processing(__SCREAMING_SNAKE_CASE , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
def _lowerCamelCase ( self : int):
'''simple docstring'''
__a = self.image_processing_class(**self.image_processor_dict)
# create random numpy tensors
__a = prepare_image_inputs(self.image_processor_tester , equal_resolution=__SCREAMING_SNAKE_CASE , numpify=__SCREAMING_SNAKE_CASE)
for image in image_inputs:
self.assertIsInstance(__SCREAMING_SNAKE_CASE , np.ndarray)
# Test not batched input
__a = image_processing(image_inputs[0] , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
__a = image_processing(__SCREAMING_SNAKE_CASE , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
def _lowerCamelCase ( self : str):
'''simple docstring'''
__a = self.image_processing_class(**self.image_processor_dict)
# create random PyTorch tensors
__a = prepare_image_inputs(self.image_processor_tester , equal_resolution=__SCREAMING_SNAKE_CASE , torchify=__SCREAMING_SNAKE_CASE)
for image in image_inputs:
self.assertIsInstance(__SCREAMING_SNAKE_CASE , torch.Tensor)
# Test not batched input
__a = image_processing(image_inputs[0] , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
1,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
# Test batched
__a = image_processing(__SCREAMING_SNAKE_CASE , return_tensors='''pt''').pixel_values
self.assertEqual(
encoded_images.shape , (
self.image_processor_tester.batch_size,
self.image_processor_tester.num_channels,
self.image_processor_tester.crop_size['''height'''],
self.image_processor_tester.crop_size['''width'''],
) , )
| 49 |
"""simple docstring"""
import datasets
__SCREAMING_SNAKE_CASE : Tuple = '\\n@InProceedings{conneau2018xnli,\n author = "Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin",\n title = "XNLI: Evaluating Cross-lingual Sentence Representations",\n booktitle = "Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing",\n year = "2018",\n publisher = "Association for Computational Linguistics",\n location = "Brussels, Belgium",\n}\n'
__SCREAMING_SNAKE_CASE : Dict = '\\nXNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n'
__SCREAMING_SNAKE_CASE : List[str] = '\nComputes XNLI score which is just simple accuracy.\nArgs:\n predictions: Predicted labels.\n references: Ground truth labels.\nReturns:\n \'accuracy\': accuracy\nExamples:\n\n >>> predictions = [0, 1]\n >>> references = [0, 1]\n >>> xnli_metric = datasets.load_metric("xnli")\n >>> results = xnli_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {\'accuracy\': 1.0}\n'
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
return (preds == labels).mean()
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION)
class __A (datasets.Metric):
'''simple docstring'''
def lowerCAmelCase ( self : str ) ->Any:
"""simple docstring"""
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
"""predictions""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
"""references""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
} ) , codebase_urls=[] , reference_urls=[] , format="""numpy""" , )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any ) ->int:
"""simple docstring"""
return {"accuracy": simple_accuracy(UpperCAmelCase_ , UpperCAmelCase_ )}
| 347 | 0 |
import argparse
import gdown
import numpy as np
import torch
from huggingface_hub import hf_hub_download
from transformers import (
CLIPTokenizer,
CLIPTokenizerFast,
VideoMAEImageProcessor,
XCLIPConfig,
XCLIPModel,
XCLIPProcessor,
XCLIPTextConfig,
XCLIPVisionConfig,
)
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase ) -> Tuple:
lowerCamelCase__ : Optional[Any] = XCLIPTextConfig()
# derive patch size from model name
lowerCamelCase__ : Optional[Any] = model_name.find('patch' )
lowerCamelCase__ : Tuple = int(model_name[start_idx + len('patch' ) : start_idx + len('patch' ) + 2] )
lowerCamelCase__ : str = XCLIPVisionConfig(patch_size=_UpperCAmelCase , num_frames=_UpperCAmelCase )
if "large" in model_name:
lowerCamelCase__ : int = 768
lowerCamelCase__ : Dict = 3072
lowerCamelCase__ : Optional[Any] = 12
lowerCamelCase__ : List[str] = 1024
lowerCamelCase__ : str = 4096
lowerCamelCase__ : Dict = 16
lowerCamelCase__ : int = 24
lowerCamelCase__ : List[Any] = 768
lowerCamelCase__ : int = 3072
if model_name == "xclip-large-patch14-16-frames":
lowerCamelCase__ : Optional[Any] = 336
lowerCamelCase__ : List[Any] = XCLIPConfig.from_text_vision_configs(_UpperCAmelCase , _UpperCAmelCase )
if "large" in model_name:
lowerCamelCase__ : List[str] = 768
return config
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> Any:
# text encoder
if name == "token_embedding.weight":
lowerCamelCase__ : int = name.replace('token_embedding.weight' , 'text_model.embeddings.token_embedding.weight' )
if name == "positional_embedding":
lowerCamelCase__ : List[Any] = name.replace('positional_embedding' , 'text_model.embeddings.position_embedding.weight' )
if "ln_1" in name:
lowerCamelCase__ : Dict = name.replace('ln_1' , 'layer_norm1' )
if "ln_2" in name:
lowerCamelCase__ : Any = name.replace('ln_2' , 'layer_norm2' )
if "c_fc" in name:
lowerCamelCase__ : str = name.replace('c_fc' , 'fc1' )
if "c_proj" in name:
lowerCamelCase__ : int = name.replace('c_proj' , 'fc2' )
if name.startswith('transformer.resblocks' ):
lowerCamelCase__ : Any = name.replace('transformer.resblocks' , 'text_model.encoder.layers' )
if "attn.out_proj" in name and "message" not in name:
lowerCamelCase__ : Union[str, Any] = name.replace('attn.out_proj' , 'self_attn.out_proj' )
if "ln_final" in name:
lowerCamelCase__ : List[Any] = name.replace('ln_final' , 'text_model.final_layer_norm' )
# visual encoder
if name == "visual.class_embedding":
lowerCamelCase__ : List[str] = name.replace('visual.class_embedding' , 'vision_model.embeddings.class_embedding' )
if name == "visual.positional_embedding":
lowerCamelCase__ : Tuple = name.replace('visual.positional_embedding' , 'vision_model.embeddings.position_embedding.weight' )
if name.startswith('visual.transformer.resblocks' ):
lowerCamelCase__ : List[Any] = name.replace('visual.transformer.resblocks' , 'vision_model.encoder.layers' )
if "visual.conv1" in name:
lowerCamelCase__ : str = name.replace('visual.conv1' , 'vision_model.embeddings.patch_embedding' )
if "visual.ln_pre" in name:
lowerCamelCase__ : Tuple = name.replace('visual.ln_pre' , 'vision_model.pre_layernorm' )
if "visual.ln_post" in name:
lowerCamelCase__ : List[str] = name.replace('visual.ln_post' , 'vision_model.post_layernorm' )
if "visual.proj" in name:
lowerCamelCase__ : str = name.replace('visual.proj' , 'visual_projection.weight' )
if "text_projection" in name:
lowerCamelCase__ : Optional[Any] = name.replace('text_projection' , 'text_projection.weight' )
# things on top
if "prompts_visual_proj" in name:
lowerCamelCase__ : Union[str, Any] = name.replace('prompts_visual_proj' , 'prompts_visual_projection' )
if "prompts_visual_ln" in name:
lowerCamelCase__ : Any = name.replace('prompts_visual_ln' , 'prompts_visual_layernorm' )
# mit
if name == "mit.positional_embedding":
lowerCamelCase__ : Any = name.replace('positional' , 'position' )
if name.startswith('mit.resblocks' ):
lowerCamelCase__ : Optional[Any] = name.replace('mit.resblocks' , 'mit.encoder.layers' )
# prompts generator
if name.startswith('prompts_generator.norm' ):
lowerCamelCase__ : Tuple = name.replace('prompts_generator.norm' , 'prompts_generator.layernorm' )
return name
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase ) -> Optional[Any]:
for key in orig_state_dict.copy().keys():
lowerCamelCase__ : Tuple = orig_state_dict.pop(_UpperCAmelCase )
if "attn.in_proj" in key:
lowerCamelCase__ : Union[str, Any] = key.split('.' )
if key.startswith('visual' ):
lowerCamelCase__ : Optional[Any] = key_split[3]
lowerCamelCase__ : Tuple = config.vision_config.hidden_size
if "message_attn" in key:
if "weight" in key:
lowerCamelCase__ : Optional[int] = val[
:dim, :
]
lowerCamelCase__ : Optional[Any] = val[
dim : dim * 2, :
]
lowerCamelCase__ : int = val[
-dim:, :
]
else:
lowerCamelCase__ : List[Any] = val[
:dim
]
lowerCamelCase__ : Optional[Any] = val[
dim : dim * 2
]
lowerCamelCase__ : Optional[int] = val[
-dim:
]
else:
if "weight" in key:
lowerCamelCase__ : int = val[
:dim, :
]
lowerCamelCase__ : Dict = val[
dim : dim * 2, :
]
lowerCamelCase__ : int = val[
-dim:, :
]
else:
lowerCamelCase__ : Optional[int] = val[:dim]
lowerCamelCase__ : str = val[
dim : dim * 2
]
lowerCamelCase__ : Optional[Any] = val[-dim:]
elif key.startswith('mit' ):
lowerCamelCase__ : List[str] = key_split[2]
lowerCamelCase__ : int = config.vision_config.mit_hidden_size
if "weight" in key:
lowerCamelCase__ : List[str] = val[:dim, :]
lowerCamelCase__ : str = val[dim : dim * 2, :]
lowerCamelCase__ : Union[str, Any] = val[-dim:, :]
else:
lowerCamelCase__ : Optional[Any] = val[:dim]
lowerCamelCase__ : List[str] = val[dim : dim * 2]
lowerCamelCase__ : Any = val[-dim:]
else:
lowerCamelCase__ : Optional[int] = key_split[2]
lowerCamelCase__ : Dict = config.text_config.hidden_size
if "weight" in key:
lowerCamelCase__ : Optional[int] = val[:dim, :]
lowerCamelCase__ : int = val[
dim : dim * 2, :
]
lowerCamelCase__ : Any = val[-dim:, :]
else:
lowerCamelCase__ : Any = val[:dim]
lowerCamelCase__ : List[Any] = val[
dim : dim * 2
]
lowerCamelCase__ : str = val[-dim:]
else:
lowerCamelCase__ : Optional[int] = rename_key(_UpperCAmelCase )
if new_key_name in ["visual_projection.weight", "text_projection.weight"]:
lowerCamelCase__ : List[Any] = val.T
lowerCamelCase__ : Optional[Any] = val
return orig_state_dict
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase ) -> Tuple:
if num_frames == 8:
lowerCamelCase__ : Any = 'eating_spaghetti_8_frames.npy'
elif num_frames == 16:
lowerCamelCase__ : Union[str, Any] = 'eating_spaghetti.npy'
elif num_frames == 32:
lowerCamelCase__ : str = 'eating_spaghetti_32_frames.npy'
lowerCamelCase__ : Dict = hf_hub_download(
repo_id='hf-internal-testing/spaghetti-video' , filename=_UpperCAmelCase , repo_type='dataset' , )
lowerCamelCase__ : str = np.load(_UpperCAmelCase )
return list(_UpperCAmelCase )
def SCREAMING_SNAKE_CASE ( _UpperCAmelCase , _UpperCAmelCase=None , _UpperCAmelCase=False ) -> Union[str, Any]:
lowerCamelCase__ : Any = {
# fully supervised kinetics-400 checkpoints
'xclip-base-patch32': 'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_8.pth',
'xclip-base-patch32-16-frames': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_32_16.pth'
),
'xclip-base-patch16': 'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_8.pth',
'xclip-base-patch16-16-frames': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k400_16_16.pth'
),
'xclip-large-patch14': 'https://drive.google.com/u/0/uc?id=1NUOImq0o5DlQTST17iIP3vG7DgmHQuCx&export=download&confirm=t&uuid=b26caedc-88e2-473e-830a-9d158b653cdb',
'xclip-large-patch14-16-frames': 'https://drive.google.com/u/0/uc?id=1FOYgnJc097OJ4lGwtRCCydQyVPJEOH7d&export=download&confirm=t&uuid=538fa810-e671-4050-b385-9a623f89804f',
# fully supervised kinetics-600 checkpoints
'xclip-base-patch16-kinetics-600': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_8.pth'
),
'xclip-base-patch16-kinetics-600-16-frames': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/k600_16_16.pth'
),
'xclip-large-patch14-kinetics-600': 'https://drive.google.com/u/0/uc?id=1FV8C1INuM91sLAN4ImjzePLIlpMSihwV&export=download&confirm=t&uuid=141d4977-4a65-44ae-864f-4b0c19f838be',
# few shot
'xclip-base-patch16-hmdb-2-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_2.pth'
),
'xclip-base-patch16-hmdb-4-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_4.pth'
),
'xclip-base-patch16-hmdb-8-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_8.pth'
),
'xclip-base-patch16-hmdb-16-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_hmdb_16.pth'
),
'xclip-base-patch16-ucf-2-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_2.pth'
),
'xclip-base-patch16-ucf-4-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_4.pth'
),
'xclip-base-patch16-ucf-8-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_8.pth'
),
'xclip-base-patch16-ucf-16-shot': (
'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/few_ucf_16.pth'
),
# zero shot
'xclip-base-patch16-zero-shot': 'https://github.com/nbl97/X-CLIP_Model_Zoo/releases/download/v1.0/zero.pth',
}
lowerCamelCase__ : List[Any] = model_to_url[model_name]
lowerCamelCase__ : Optional[Any] = 8
if "16-frames" in model_name:
lowerCamelCase__ : Union[str, Any] = 16
elif "shot" in model_name:
lowerCamelCase__ : Optional[int] = 32
lowerCamelCase__ : Tuple = get_xclip_config(_UpperCAmelCase , _UpperCAmelCase )
lowerCamelCase__ : List[Any] = XCLIPModel(_UpperCAmelCase )
model.eval()
if "drive" in checkpoint_url:
lowerCamelCase__ : Dict = 'pytorch_model.bin'
gdown.cached_download(_UpperCAmelCase , _UpperCAmelCase , quiet=_UpperCAmelCase )
lowerCamelCase__ : Any = torch.load(_UpperCAmelCase , map_location='cpu' )['model']
else:
lowerCamelCase__ : Tuple = torch.hub.load_state_dict_from_url(_UpperCAmelCase )['model']
lowerCamelCase__ : Any = convert_state_dict(_UpperCAmelCase , _UpperCAmelCase )
lowerCamelCase__ : List[Any] = XCLIPModel(_UpperCAmelCase )
lowerCamelCase__ , lowerCamelCase__ : List[Any] = model.load_state_dict(_UpperCAmelCase , strict=_UpperCAmelCase )
assert missing_keys == ["text_model.embeddings.position_ids", "vision_model.embeddings.position_ids"]
model.eval()
lowerCamelCase__ : Any = 336 if model_name == 'xclip-large-patch14-16-frames' else 224
lowerCamelCase__ : List[str] = VideoMAEImageProcessor(size=_UpperCAmelCase )
lowerCamelCase__ : str = CLIPTokenizer.from_pretrained('openai/clip-vit-base-patch32' )
lowerCamelCase__ : List[Any] = CLIPTokenizerFast.from_pretrained('openai/clip-vit-base-patch32' )
lowerCamelCase__ : Optional[int] = XCLIPProcessor(image_processor=_UpperCAmelCase , tokenizer=_UpperCAmelCase )
lowerCamelCase__ : int = prepare_video(_UpperCAmelCase )
lowerCamelCase__ : Tuple = processor(
text=['playing sports', 'eating spaghetti', 'go shopping'] , videos=_UpperCAmelCase , return_tensors='pt' , padding=_UpperCAmelCase )
print('Shape of pixel values:' , inputs.pixel_values.shape )
with torch.no_grad():
lowerCamelCase__ : List[Any] = model(**_UpperCAmelCase )
# Verify outputs
lowerCamelCase__ : str = outputs.logits_per_video
lowerCamelCase__ : Union[str, Any] = logits_per_video.softmax(dim=1 )
print('Probs:' , _UpperCAmelCase )
# kinetics-400
if model_name == "xclip-base-patch32":
lowerCamelCase__ : Optional[Any] = torch.tensor([[0.0_019, 0.9_951, 0.0_030]] )
elif model_name == "xclip-base-patch32-16-frames":
lowerCamelCase__ : Union[str, Any] = torch.tensor([[7.0_999e-04, 9.9_883e-01, 4.5_580e-04]] )
elif model_name == "xclip-base-patch16":
lowerCamelCase__ : int = torch.tensor([[0.0_083, 0.9_681, 0.0_236]] )
elif model_name == "xclip-base-patch16-16-frames":
lowerCamelCase__ : Any = torch.tensor([[7.6_937e-04, 9.9_728e-01, 1.9_473e-03]] )
elif model_name == "xclip-large-patch14":
lowerCamelCase__ : Tuple = torch.tensor([[0.0_062, 0.9_864, 0.0_075]] )
elif model_name == "xclip-large-patch14-16-frames":
lowerCamelCase__ : Optional[int] = torch.tensor([[3.3_877e-04, 9.9_937e-01, 2.8_888e-04]] )
# kinetics-600
elif model_name == "xclip-base-patch16-kinetics-600":
lowerCamelCase__ : Tuple = torch.tensor([[0.0_555, 0.8_914, 0.0_531]] )
elif model_name == "xclip-base-patch16-kinetics-600-16-frames":
lowerCamelCase__ : Tuple = torch.tensor([[3.8_554e-04, 9.9_929e-01, 3.2_754e-04]] )
elif model_name == "xclip-large-patch14-kinetics-600":
lowerCamelCase__ : List[Any] = torch.tensor([[0.0_036, 0.9_920, 0.0_045]] )
# few shot
elif model_name == "xclip-base-patch16-hmdb-2-shot":
lowerCamelCase__ : Tuple = torch.tensor([[7.1_890e-06, 9.9_994e-01, 5.6_559e-05]] )
elif model_name == "xclip-base-patch16-hmdb-4-shot":
lowerCamelCase__ : Any = torch.tensor([[1.0_320e-05, 9.9_993e-01, 6.2_435e-05]] )
elif model_name == "xclip-base-patch16-hmdb-8-shot":
lowerCamelCase__ : Optional[Any] = torch.tensor([[4.1_377e-06, 9.9_990e-01, 9.8_386e-05]] )
elif model_name == "xclip-base-patch16-hmdb-16-shot":
lowerCamelCase__ : Optional[int] = torch.tensor([[4.1_347e-05, 9.9_962e-01, 3.3_411e-04]] )
elif model_name == "xclip-base-patch16-ucf-2-shot":
lowerCamelCase__ : int = torch.tensor([[8.5_857e-05, 9.9_928e-01, 6.3_291e-04]] )
elif model_name == "xclip-base-patch16-ucf-4-shot":
lowerCamelCase__ : Dict = torch.tensor([[8.5_857e-05, 9.9_928e-01, 6.3_291e-04]] )
elif model_name == "xclip-base-patch16-ucf-8-shot":
lowerCamelCase__ : List[Any] = torch.tensor([[0.0_027, 0.9_904, 0.0_070]] )
elif model_name == "xclip-base-patch16-ucf-16-shot":
lowerCamelCase__ : Union[str, Any] = torch.tensor([[9.8_219e-04, 9.9_593e-01, 3.0_863e-03]] )
# zero shot
elif model_name == "xclip-base-patch16-zero-shot":
lowerCamelCase__ : int = torch.tensor([[3.5_082e-04, 9.9_785e-01, 1.7_966e-03]] )
else:
raise ValueError(F"""Model name {model_name} not supported""" )
assert torch.allclose(_UpperCAmelCase , _UpperCAmelCase , atol=1e-3 )
print('Looks ok!' )
if pytorch_dump_folder_path is not None:
print(F"""Saving model {model_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_UpperCAmelCase )
if push_to_hub:
print('Pushing model, processor and slow tokenizer files to the hub...' )
model.push_to_hub(_UpperCAmelCase , organization='nielsr' )
processor.push_to_hub(_UpperCAmelCase , organization='nielsr' )
slow_tokenizer.push_to_hub(_UpperCAmelCase , organization='nielsr' )
if __name__ == "__main__":
_UpperCAmelCase : List[str] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--model_name""",
default="""xclip-base-patch32""",
type=str,
help="""Name of the model.""",
)
parser.add_argument(
"""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model directory."""
)
parser.add_argument(
"""--push_to_hub""", action="""store_true""", help="""Whether or not to push the converted model to the 🤗 hub."""
)
_UpperCAmelCase : List[str] = parser.parse_args()
convert_xclip_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
| 50 |
"""simple docstring"""
from ..utils import DummyObject, requires_backends
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Tuple ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : int ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Dict ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Union[str, Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[Any] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : str , **UpperCAmelCase_ : int ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : Union[str, Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[Any] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : List[str] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Optional[Any] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Optional[Any] ) ->int:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Dict ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Union[str, Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[int] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Optional[int] ) ->Any:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
| 347 | 0 |
from collections.abc import Iterable
from typing import Any
class __snake_case :
def __init__( self : Tuple , _snake_case : int | None = None):
"""simple docstring"""
UpperCAmelCase_ = value
UpperCAmelCase_ = None # Added in order to delete a node easier
UpperCAmelCase_ = None
UpperCAmelCase_ = None
def __repr__( self : Dict):
"""simple docstring"""
from pprint import pformat
if self.left is None and self.right is None:
return str(self.value)
return pformat({F"""{self.value}""": (self.left, self.right)} , indent=1)
class __snake_case :
def __init__( self : int , _snake_case : Node | None = None):
"""simple docstring"""
UpperCAmelCase_ = root
def __str__( self : List[str]):
"""simple docstring"""
return str(self.root)
def lowerCamelCase ( self : Any , _snake_case : Node , _snake_case : Node | None):
"""simple docstring"""
if new_children is not None: # reset its kids
UpperCAmelCase_ = node.parent
if node.parent is not None: # reset its parent
if self.is_right(_snake_case): # If it is the right children
UpperCAmelCase_ = new_children
else:
UpperCAmelCase_ = new_children
else:
UpperCAmelCase_ = new_children
def lowerCamelCase ( self : int , _snake_case : Node):
"""simple docstring"""
if node.parent and node.parent.right:
return node == node.parent.right
return False
def lowerCamelCase ( self : Optional[Any]):
"""simple docstring"""
return self.root is None
def lowerCamelCase ( self : Any , _snake_case : str):
"""simple docstring"""
UpperCAmelCase_ = Node(_snake_case) # create a new Node
if self.empty(): # if Tree is empty
UpperCAmelCase_ = new_node # set its root
else: # Tree is not empty
UpperCAmelCase_ = self.root # from root
if parent_node is None:
return
while True: # While we don't get to a leaf
if value < parent_node.value: # We go left
if parent_node.left is None:
UpperCAmelCase_ = new_node # We insert the new node in a leaf
break
else:
UpperCAmelCase_ = parent_node.left
else:
if parent_node.right is None:
UpperCAmelCase_ = new_node
break
else:
UpperCAmelCase_ = parent_node.right
UpperCAmelCase_ = parent_node
def lowerCamelCase ( self : Any , *_snake_case : Union[str, Any]):
"""simple docstring"""
for value in values:
self.__insert(_snake_case)
def lowerCamelCase ( self : Any , _snake_case : List[str]):
"""simple docstring"""
if self.empty():
raise IndexError('''Warning: Tree is empty! please use another.''')
else:
UpperCAmelCase_ = self.root
# use lazy evaluation here to avoid NoneType Attribute error
while node is not None and node.value is not value:
UpperCAmelCase_ = node.left if value < node.value else node.right
return node
def lowerCamelCase ( self : List[Any] , _snake_case : Node | None = None):
"""simple docstring"""
if node is None:
if self.root is None:
return None
UpperCAmelCase_ = self.root
if not self.empty():
while node.right is not None:
UpperCAmelCase_ = node.right
return node
def lowerCamelCase ( self : Any , _snake_case : Node | None = None):
"""simple docstring"""
if node is None:
UpperCAmelCase_ = self.root
if self.root is None:
return None
if not self.empty():
UpperCAmelCase_ = self.root
while node.left is not None:
UpperCAmelCase_ = node.left
return node
def lowerCamelCase ( self : Dict , _snake_case : int):
"""simple docstring"""
UpperCAmelCase_ = self.search(_snake_case) # Look for the node with that label
if node is not None:
if node.left is None and node.right is None: # If it has no children
self.__reassign_nodes(_snake_case , _snake_case)
elif node.left is None: # Has only right children
self.__reassign_nodes(_snake_case , node.right)
elif node.right is None: # Has only left children
self.__reassign_nodes(_snake_case , node.left)
else:
UpperCAmelCase_ = self.get_max(
node.left) # Gets the max value of the left branch
self.remove(tmp_node.value) # type: ignore
UpperCAmelCase_ = (
tmp_node.value # type: ignore
) # Assigns the value to the node to delete and keep tree structure
def lowerCamelCase ( self : Tuple , _snake_case : Node | None):
"""simple docstring"""
if node is not None:
yield node # Preorder Traversal
yield from self.preorder_traverse(node.left)
yield from self.preorder_traverse(node.right)
def lowerCamelCase ( self : Optional[int] , _snake_case : str=None):
"""simple docstring"""
if traversal_function is None:
return self.preorder_traverse(self.root)
else:
return traversal_function(self.root)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : list , _snake_case : Node | None):
"""simple docstring"""
if node:
self.inorder(_snake_case , node.left)
arr.append(node.value)
self.inorder(_snake_case , node.right)
def lowerCamelCase ( self : Union[str, Any] , _snake_case : int , _snake_case : Node):
"""simple docstring"""
UpperCAmelCase_ = []
self.inorder(_snake_case , _snake_case) # append all values to list using inorder traversal
return arr[k - 1]
def A (__A : Node | None ) -> list[Node]:
"""simple docstring"""
UpperCAmelCase_ = []
if curr_node is not None:
UpperCAmelCase_ = postorder(curr_node.left ) + postorder(curr_node.right ) + [curr_node]
return node_list
def A () -> None:
"""simple docstring"""
UpperCAmelCase_ = (8, 3, 6, 1, 10, 14, 13, 4, 7)
UpperCAmelCase_ = BinarySearchTree()
for i in testlist:
t.insert(__A )
# Prints all the elements of the list in order traversal
print(__A )
if t.search(6 ) is not None:
print('''The value 6 exists''' )
else:
print('''The value 6 doesn\'t exist''' )
if t.search(-1 ) is not None:
print('''The value -1 exists''' )
else:
print('''The value -1 doesn\'t exist''' )
if not t.empty():
print('''Max Value: ''' , t.get_max().value ) # type: ignore
print('''Min Value: ''' , t.get_min().value ) # type: ignore
for i in testlist:
t.remove(__A )
print(__A )
if __name__ == "__main__":
import doctest
doctest.testmod(verbose=True)
| 51 |
"""simple docstring"""
import warnings
from ...utils import logging
from .image_processing_mobilevit import MobileViTImageProcessor
__SCREAMING_SNAKE_CASE : int = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
def __init__( self : str , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->None:
"""simple docstring"""
warnings.warn(
"""The class MobileViTFeatureExtractor is deprecated and will be removed in version 5 of Transformers."""
""" Please use MobileViTImageProcessor instead.""" , UpperCAmelCase_ , )
super().__init__(*UpperCAmelCase_ , **UpperCAmelCase_ )
| 347 | 0 |
import os
import tempfile
from functools import partial
from unittest import TestCase
from unittest.mock import patch
import numpy as np
import pytest
from datasets.arrow_dataset import Dataset
from datasets.search import ElasticSearchIndex, FaissIndex, MissingIndex
from .utils import require_elasticsearch, require_faiss
__lowerCamelCase : Union[str, Any] = pytest.mark.integration
@require_faiss
class A__ ( __snake_case ):
def __UpperCamelCase( self ):
'''simple docstring'''
UpperCamelCase : Union[str, Any] = Dataset.from_dict({"filename": ["my_name-train" + "_" + str(A_ ) for x in np.arange(30 ).tolist()]} )
return dset
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : Dataset = self._create_dummy_dataset()
UpperCamelCase : List[Any] = dset.map(
lambda A_ , A_ : {"vecs": i * np.ones(5 , dtype=np.floataa )} , with_indices=A_ , keep_in_memory=A_ )
UpperCamelCase : List[str] = dset.add_faiss_index("vecs" , batch_size=100 , metric_type=faiss.METRIC_INNER_PRODUCT )
UpperCamelCase , UpperCamelCase : Tuple = dset.get_nearest_examples("vecs" , np.ones(5 , dtype=np.floataa ) )
self.assertEqual(examples["filename"][0] , "my_name-train_29" )
dset.drop_index("vecs" )
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : Dataset = self._create_dummy_dataset()
dset.add_faiss_index_from_external_arrays(
external_arrays=np.ones((30, 5) ) * np.arange(30 ).reshape(-1 , 1 ) , index_name="vecs" , batch_size=100 , metric_type=faiss.METRIC_INNER_PRODUCT , )
UpperCamelCase , UpperCamelCase : int = dset.get_nearest_examples("vecs" , np.ones(5 , dtype=np.floataa ) )
self.assertEqual(examples["filename"][0] , "my_name-train_29" )
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : Dataset = self._create_dummy_dataset()
dset.add_faiss_index_from_external_arrays(
external_arrays=np.ones((30, 5) ) * np.arange(30 ).reshape(-1 , 1 ) , index_name="vecs" , metric_type=faiss.METRIC_INNER_PRODUCT , )
# Setting delete=False and unlinking manually is not pretty... but it is required on Windows to
# ensure somewhat stable behaviour. If we don't, we get PermissionErrors. This is an age-old issue.
# see https://bugs.python.org/issue14243 and
# https://stackoverflow.com/questions/23212435/permission-denied-to-write-to-my-temporary-file/23212515
with tempfile.NamedTemporaryFile(delete=A_ ) as tmp_file:
dset.save_faiss_index("vecs" , tmp_file.name )
dset.load_faiss_index("vecs2" , tmp_file.name )
os.unlink(tmp_file.name )
UpperCamelCase , UpperCamelCase : List[str] = dset.get_nearest_examples("vecs2" , np.ones(5 , dtype=np.floataa ) )
self.assertEqual(examples["filename"][0] , "my_name-train_29" )
def __UpperCamelCase( self ):
'''simple docstring'''
UpperCamelCase : Dataset = self._create_dummy_dataset()
dset.add_faiss_index_from_external_arrays(
external_arrays=np.ones((30, 5) ) * np.arange(30 ).reshape(-1 , 1 ) , index_name="vecs" )
dset.drop_index("vecs" )
self.assertRaises(A_ , partial(dset.get_nearest_examples , "vecs2" , np.ones(5 , dtype=np.floataa ) ) )
def __UpperCamelCase( self ):
'''simple docstring'''
from elasticsearch import Elasticsearch
UpperCamelCase : Dataset = self._create_dummy_dataset()
with patch("elasticsearch.Elasticsearch.search" ) as mocked_search, patch(
"elasticsearch.client.IndicesClient.create" ) as mocked_index_create, patch("elasticsearch.helpers.streaming_bulk" ) as mocked_bulk:
UpperCamelCase : List[str] = {"acknowledged": True}
mocked_bulk.return_value([(True, None)] * 30 )
UpperCamelCase : List[Any] = {"hits": {"hits": [{"_score": 1, "_id": 29}]}}
UpperCamelCase : Optional[Any] = Elasticsearch()
dset.add_elasticsearch_index("filename" , es_client=A_ )
UpperCamelCase , UpperCamelCase : List[str] = dset.get_nearest_examples("filename" , "my_name-train_29" )
self.assertEqual(examples["filename"][0] , "my_name-train_29" )
@require_faiss
class A__ ( __snake_case ):
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : Optional[int] = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT )
# add vectors
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
self.assertIsNotNone(index.faiss_index )
self.assertEqual(index.faiss_index.ntotal , 5 )
index.add_vectors(np.zeros((5, 5) , dtype=np.floataa ) )
self.assertEqual(index.faiss_index.ntotal , 10 )
# single query
UpperCamelCase : Any = np.zeros(5 , dtype=np.floataa )
UpperCamelCase : Optional[Any] = 1
UpperCamelCase , UpperCamelCase : Optional[Any] = index.search(A_ )
self.assertRaises(A_ , index.search , query.reshape(-1 , 1 ) )
self.assertGreater(scores[0] , 0 )
self.assertEqual(indices[0] , 1 )
# batched queries
UpperCamelCase : Optional[int] = np.eye(5 , dtype=np.floataa )[::-1]
UpperCamelCase , UpperCamelCase : Tuple = index.search_batch(A_ )
self.assertRaises(A_ , index.search_batch , queries[0] )
UpperCamelCase : Optional[int] = [scores[0] for scores in total_scores]
UpperCamelCase : Tuple = [indices[0] for indices in total_indices]
self.assertGreater(np.min(A_ ) , 0 )
self.assertListEqual([4, 3, 2, 1, 0] , A_ )
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : List[str] = FaissIndex(string_factory="Flat" )
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
self.assertIsInstance(index.faiss_index , faiss.IndexFlat )
UpperCamelCase : List[str] = FaissIndex(string_factory="LSH" )
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
self.assertIsInstance(index.faiss_index , faiss.IndexLSH )
with self.assertRaises(A_ ):
UpperCamelCase : List[str] = FaissIndex(string_factory="Flat" , custom_index=faiss.IndexFlat(5 ) )
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : Dict = faiss.IndexFlat(5 )
UpperCamelCase : Union[str, Any] = FaissIndex(custom_index=A_ )
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
self.assertIsInstance(index.faiss_index , faiss.IndexFlat )
def __UpperCamelCase( self ):
'''simple docstring'''
import faiss
UpperCamelCase : str = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT )
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
# Setting delete=False and unlinking manually is not pretty... but it is required on Windows to
# ensure somewhat stable behaviour. If we don't, we get PermissionErrors. This is an age-old issue.
# see https://bugs.python.org/issue14243 and
# https://stackoverflow.com/questions/23212435/permission-denied-to-write-to-my-temporary-file/23212515
with tempfile.NamedTemporaryFile(delete=A_ ) as tmp_file:
index.save(tmp_file.name )
UpperCamelCase : int = FaissIndex.load(tmp_file.name )
os.unlink(tmp_file.name )
UpperCamelCase : str = np.zeros(5 , dtype=np.floataa )
UpperCamelCase : int = 1
UpperCamelCase , UpperCamelCase : Dict = index.search(A_ )
self.assertGreater(scores[0] , 0 )
self.assertEqual(indices[0] , 1 )
@require_faiss
def A_ ( _lowerCAmelCase ) -> Optional[int]:
import faiss
UpperCamelCase : Union[str, Any] = FaissIndex(metric_type=faiss.METRIC_INNER_PRODUCT )
index.add_vectors(np.eye(5 , dtype=np.floataa ) )
UpperCamelCase : List[Any] = "index.faiss"
UpperCamelCase : List[str] = F"""mock://{index_name}"""
index.save(_lowerCAmelCase , storage_options=mockfs.storage_options )
UpperCamelCase : List[str] = FaissIndex.load(_lowerCAmelCase , storage_options=mockfs.storage_options )
UpperCamelCase : List[str] = np.zeros(5 , dtype=np.floataa )
UpperCamelCase : Optional[int] = 1
UpperCamelCase , UpperCamelCase : List[str] = index.search(_lowerCAmelCase )
assert scores[0] > 0
assert indices[0] == 1
@require_elasticsearch
class A__ ( __snake_case ):
def __UpperCamelCase( self ):
'''simple docstring'''
from elasticsearch import Elasticsearch
with patch("elasticsearch.Elasticsearch.search" ) as mocked_search, patch(
"elasticsearch.client.IndicesClient.create" ) as mocked_index_create, patch("elasticsearch.helpers.streaming_bulk" ) as mocked_bulk:
UpperCamelCase : List[str] = Elasticsearch()
UpperCamelCase : Union[str, Any] = {"acknowledged": True}
UpperCamelCase : Union[str, Any] = ElasticSearchIndex(es_client=A_ )
mocked_bulk.return_value([(True, None)] * 3 )
index.add_documents(["foo", "bar", "foobar"] )
# single query
UpperCamelCase : str = "foo"
UpperCamelCase : Dict = {"hits": {"hits": [{"_score": 1, "_id": 0}]}}
UpperCamelCase , UpperCamelCase : Tuple = index.search(A_ )
self.assertEqual(scores[0] , 1 )
self.assertEqual(indices[0] , 0 )
# single query with timeout
UpperCamelCase : Dict = "foo"
UpperCamelCase : Optional[Any] = {"hits": {"hits": [{"_score": 1, "_id": 0}]}}
UpperCamelCase , UpperCamelCase : str = index.search(A_ , request_timeout=30 )
self.assertEqual(scores[0] , 1 )
self.assertEqual(indices[0] , 0 )
# batched queries
UpperCamelCase : Dict = ["foo", "bar", "foobar"]
UpperCamelCase : List[Any] = {"hits": {"hits": [{"_score": 1, "_id": 1}]}}
UpperCamelCase , UpperCamelCase : Optional[int] = index.search_batch(A_ )
UpperCamelCase : str = [scores[0] for scores in total_scores]
UpperCamelCase : Optional[Any] = [indices[0] for indices in total_indices]
self.assertGreater(np.min(A_ ) , 0 )
self.assertListEqual([1, 1, 1] , A_ )
# batched queries with timeout
UpperCamelCase : int = ["foo", "bar", "foobar"]
UpperCamelCase : List[Any] = {"hits": {"hits": [{"_score": 1, "_id": 1}]}}
UpperCamelCase , UpperCamelCase : Union[str, Any] = index.search_batch(A_ , request_timeout=30 )
UpperCamelCase : Union[str, Any] = [scores[0] for scores in total_scores]
UpperCamelCase : Dict = [indices[0] for indices in total_indices]
self.assertGreater(np.min(A_ ) , 0 )
self.assertListEqual([1, 1, 1] , A_ )
| 52 |
"""simple docstring"""
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Any:
snake_case_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
("""cls_token""", """vit.embeddings.cls_token"""),
("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight"""),
("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias"""),
("""pos_embed""", """vit.embeddings.position_embeddings"""),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("""norm.weight""", """layernorm.weight"""),
("""norm.bias""", """layernorm.bias"""),
("""pre_logits.fc.weight""", """pooler.dense.weight"""),
("""pre_logits.fc.bias""", """pooler.dense.bias"""),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
snake_case_ = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("""norm.weight""", """vit.layernorm.weight"""),
("""norm.bias""", """vit.layernorm.bias"""),
("""head.weight""", """classifier.weight"""),
("""head.bias""", """classifier.bias"""),
] )
return rename_keys
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
snake_case_ = """"""
else:
snake_case_ = """vit."""
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case_ = in_proj_weight[
: config.hidden_size, :
]
snake_case_ = in_proj_bias[: config.hidden_size]
snake_case_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case_ = in_proj_weight[
-config.hidden_size :, :
]
snake_case_ = in_proj_bias[-config.hidden_size :]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = ["""head.weight""", """head.bias"""]
for k in ignore_keys:
state_dict.pop(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> str:
snake_case_ = dct.pop(_SCREAMING_SNAKE_CASE )
snake_case_ = val
def _a ( ) -> Any:
snake_case_ = """http://images.cocodataset.org/val2017/000000039769.jpg"""
snake_case_ = Image.open(requests.get(_SCREAMING_SNAKE_CASE , stream=_SCREAMING_SNAKE_CASE ).raw )
return im
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = ViTConfig()
snake_case_ = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
snake_case_ = True
snake_case_ = int(vit_name[-12:-10] )
snake_case_ = int(vit_name[-9:-6] )
else:
snake_case_ = 1_000
snake_case_ = """huggingface/label-files"""
snake_case_ = """imagenet-1k-id2label.json"""
snake_case_ = json.load(open(hf_hub_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , repo_type="""dataset""" ) , """r""" ) )
snake_case_ = {int(_SCREAMING_SNAKE_CASE ): v for k, v in idalabel.items()}
snake_case_ = idalabel
snake_case_ = {v: k for k, v in idalabel.items()}
snake_case_ = int(vit_name[-6:-4] )
snake_case_ = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith("""tiny""" ):
snake_case_ = 192
snake_case_ = 768
snake_case_ = 12
snake_case_ = 3
elif vit_name[9:].startswith("""small""" ):
snake_case_ = 384
snake_case_ = 1_536
snake_case_ = 12
snake_case_ = 6
else:
pass
else:
if vit_name[4:].startswith("""small""" ):
snake_case_ = 768
snake_case_ = 2_304
snake_case_ = 8
snake_case_ = 8
elif vit_name[4:].startswith("""base""" ):
pass
elif vit_name[4:].startswith("""large""" ):
snake_case_ = 1_024
snake_case_ = 4_096
snake_case_ = 24
snake_case_ = 16
elif vit_name[4:].startswith("""huge""" ):
snake_case_ = 1_280
snake_case_ = 5_120
snake_case_ = 32
snake_case_ = 16
# load original model from timm
snake_case_ = timm.create_model(_SCREAMING_SNAKE_CASE , pretrained=_SCREAMING_SNAKE_CASE )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case_ = timm_model.state_dict()
if base_model:
remove_classification_head_(_SCREAMING_SNAKE_CASE )
snake_case_ = create_rename_keys(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
for src, dest in rename_keys:
rename_key(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
read_in_q_k_v(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# load HuggingFace model
if vit_name[-5:] == "in21k":
snake_case_ = ViTModel(_SCREAMING_SNAKE_CASE ).eval()
else:
snake_case_ = ViTForImageClassification(_SCREAMING_SNAKE_CASE ).eval()
model.load_state_dict(_SCREAMING_SNAKE_CASE )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
snake_case_ = DeiTImageProcessor(size=config.image_size )
else:
snake_case_ = ViTImageProcessor(size=config.image_size )
snake_case_ = image_processor(images=prepare_img() , return_tensors="""pt""" )
snake_case_ = encoding["""pixel_values"""]
snake_case_ = model(_SCREAMING_SNAKE_CASE )
if base_model:
snake_case_ = timm_model.forward_features(_SCREAMING_SNAKE_CASE )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.pooler_output , atol=1E-3 )
else:
snake_case_ = timm_model(_SCREAMING_SNAKE_CASE )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.logits , atol=1E-3 )
Path(_SCREAMING_SNAKE_CASE ).mkdir(exist_ok=_SCREAMING_SNAKE_CASE )
print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_SCREAMING_SNAKE_CASE )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--vit_name',
default='vit_base_patch16_224',
type=str,
help='Name of the ViT timm model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
__SCREAMING_SNAKE_CASE : int = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 347 | 0 |
'''simple docstring'''
import random
import sys
import numpy as np
from matplotlib import pyplot as plt
from matplotlib.colors import ListedColormap
a__ : Union[str, Any] ='''Usage of script: script_name <size_of_canvas:int>'''
a__ : Tuple =[0] * 100 + [1] * 10
random.shuffle(choice)
def lowercase__ ( __lowercase : int ) -> list[list[bool]]:
"""simple docstring"""
__UpperCamelCase = [[False for i in range(__lowercase )] for j in range(__lowercase )]
return canvas
def lowercase__ ( __lowercase : list[list[bool]] ) -> None:
"""simple docstring"""
for i, row in enumerate(__lowercase ):
for j, _ in enumerate(__lowercase ):
__UpperCamelCase = bool(random.getrandbits(1 ) )
def lowercase__ ( __lowercase : list[list[bool]] ) -> list[list[bool]]:
"""simple docstring"""
__UpperCamelCase = np.array(__lowercase )
__UpperCamelCase = np.array(create_canvas(current_canvas.shape[0] ) )
for r, row in enumerate(__lowercase ):
for c, pt in enumerate(__lowercase ):
__UpperCamelCase = __judge_point(
__lowercase , current_canvas[r - 1 : r + 2, c - 1 : c + 2] )
__UpperCamelCase = next_gen_canvas
del next_gen_canvas # cleaning memory as we move on.
__UpperCamelCase = current_canvas.tolist()
return return_canvas
def lowercase__ ( __lowercase : bool , __lowercase : list[list[bool]] ) -> bool:
"""simple docstring"""
__UpperCamelCase = 0
__UpperCamelCase = 0
# finding dead or alive neighbours count.
for i in neighbours:
for status in i:
if status:
alive += 1
else:
dead += 1
# handling duplicate entry for focus pt.
if pt:
alive -= 1
else:
dead -= 1
# running the rules of game here.
__UpperCamelCase = pt
if pt:
if alive < 2:
__UpperCamelCase = False
elif alive == 2 or alive == 3:
__UpperCamelCase = True
elif alive > 3:
__UpperCamelCase = False
else:
if alive == 3:
__UpperCamelCase = True
return state
if __name__ == "__main__":
if len(sys.argv) != 2:
raise Exception(usage_doc)
a__ : Union[str, Any] =int(sys.argv[1])
# main working structure of this module.
a__ : Any =create_canvas(canvas_size)
seed(c)
a__ , a__ : Union[str, Any] =plt.subplots()
fig.show()
a__ : Tuple =ListedColormap(['''w''', '''k'''])
try:
while True:
a__ : int =run(c)
ax.matshow(c, cmap=cmap)
fig.canvas.draw()
ax.cla()
except KeyboardInterrupt:
# do nothing.
pass
| 53 |
"""simple docstring"""
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A (unittest.TestCase):
'''simple docstring'''
def __init__( self : List[Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Tuple=13 , UpperCAmelCase_ : List[Any]=7 , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=99 , UpperCAmelCase_ : str=32 , UpperCAmelCase_ : Tuple=5 , UpperCAmelCase_ : Union[str, Any]=4 , UpperCAmelCase_ : Any=37 , UpperCAmelCase_ : int="gelu" , UpperCAmelCase_ : Any=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Dict=512 , UpperCAmelCase_ : Optional[Any]=16 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : str=0.02 , UpperCAmelCase_ : str=4 , ) ->Tuple:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_attention_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = type_vocab_size
snake_case_ = type_sequence_label_size
snake_case_ = initializer_range
snake_case_ = num_choices
def lowerCAmelCase ( self : Optional[int] ) ->str:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_attention_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = None
if self.use_token_type_ids:
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=UpperCAmelCase_ , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def lowerCAmelCase ( self : List[str] ) ->Dict:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """token_type_ids""": token_type_ids, """attention_mask""": attention_mask}
return config, inputs_dict
@require_flax
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Union[str, Any] = True
__lowercase: int = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = FlaxRoFormerModelTester(self )
@slow
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""junnyu/roformer_chinese_small""" , from_pt=UpperCAmelCase_ )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
@require_flax
class __A (unittest.TestCase):
'''simple docstring'''
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = FlaxRoFormerForMaskedLM.from_pretrained("""junnyu/roformer_chinese_base""" )
snake_case_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
snake_case_ = model(UpperCAmelCase_ )[0]
snake_case_ = 50_000
snake_case_ = (1, 6, vocab_size)
self.assertEqual(output.shape , UpperCAmelCase_ )
snake_case_ = jnp.array(
[[[-0.1_205, -1.0_265, 0.2_922], [-1.5_134, 0.1_974, 0.1_519], [-5.0_135, -3.9_003, -0.8_404]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=1E-4 ) )
| 347 | 0 |
"""simple docstring"""
import unittest
from accelerate import debug_launcher
from accelerate.test_utils import require_cpu, test_ops, test_script
@require_cpu
class UpperCamelCase_ ( unittest.TestCase):
"""simple docstring"""
def UpperCAmelCase_ ( self : List[str] ) -> Optional[Any]:
debug_launcher(test_script.main )
def UpperCAmelCase_ ( self : List[Any] ) -> Any:
debug_launcher(test_ops.main )
| 54 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> bool:
snake_case_ = get_failure_array(_SCREAMING_SNAKE_CASE )
# 2) Step through text searching for pattern
snake_case_ , snake_case_ = 0, 0 # index into text, pattern
while i < len(_SCREAMING_SNAKE_CASE ):
if pattern[j] == text[i]:
if j == (len(_SCREAMING_SNAKE_CASE ) - 1):
return True
j += 1
# if this is a prefix in our pattern
# just go back far enough to continue
elif j > 0:
snake_case_ = failure[j - 1]
continue
i += 1
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> list[int]:
snake_case_ = [0]
snake_case_ = 0
snake_case_ = 1
while j < len(_SCREAMING_SNAKE_CASE ):
if pattern[i] == pattern[j]:
i += 1
elif i > 0:
snake_case_ = failure[i - 1]
continue
j += 1
failure.append(_SCREAMING_SNAKE_CASE )
return failure
if __name__ == "__main__":
# Test 1)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abc1abc12'
__SCREAMING_SNAKE_CASE : Optional[int] = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
__SCREAMING_SNAKE_CASE : List[str] = 'alskfjaldsk23adsfabcabc'
assert kmp(pattern, texta) and not kmp(pattern, texta)
# Test 2)
__SCREAMING_SNAKE_CASE : int = 'ABABX'
__SCREAMING_SNAKE_CASE : Optional[Any] = 'ABABZABABYABABX'
assert kmp(pattern, text)
# Test 3)
__SCREAMING_SNAKE_CASE : Any = 'AAAB'
__SCREAMING_SNAKE_CASE : List[Any] = 'ABAAAAAB'
assert kmp(pattern, text)
# Test 4)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abcdabcy'
__SCREAMING_SNAKE_CASE : str = 'abcxabcdabxabcdabcdabcy'
assert kmp(pattern, text)
# Test 5)
__SCREAMING_SNAKE_CASE : Any = 'aabaabaaa'
assert get_failure_array(pattern) == [0, 1, 0, 1, 2, 3, 4, 5, 2]
| 347 | 0 |
'''simple docstring'''
import argparse
import json
from pathlib import Path
import requests
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
a_ : Tuple = logging.get_logger(__name__)
def __snake_case ( UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : List[str]=False ):
lowerCamelCase_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((F'''blocks.{i}.norm1.weight''', F'''vit.encoder.layer.{i}.layernorm_before.weight''') )
rename_keys.append((F'''blocks.{i}.norm1.bias''', F'''vit.encoder.layer.{i}.layernorm_before.bias''') )
rename_keys.append((F'''blocks.{i}.attn.proj.weight''', F'''vit.encoder.layer.{i}.attention.output.dense.weight''') )
rename_keys.append((F'''blocks.{i}.attn.proj.bias''', F'''vit.encoder.layer.{i}.attention.output.dense.bias''') )
rename_keys.append((F'''blocks.{i}.norm2.weight''', F'''vit.encoder.layer.{i}.layernorm_after.weight''') )
rename_keys.append((F'''blocks.{i}.norm2.bias''', F'''vit.encoder.layer.{i}.layernorm_after.bias''') )
rename_keys.append((F'''blocks.{i}.mlp.fc1.weight''', F'''vit.encoder.layer.{i}.intermediate.dense.weight''') )
rename_keys.append((F'''blocks.{i}.mlp.fc1.bias''', F'''vit.encoder.layer.{i}.intermediate.dense.bias''') )
rename_keys.append((F'''blocks.{i}.mlp.fc2.weight''', F'''vit.encoder.layer.{i}.output.dense.weight''') )
rename_keys.append((F'''blocks.{i}.mlp.fc2.bias''', F'''vit.encoder.layer.{i}.output.dense.bias''') )
# projection layer + position embeddings
rename_keys.extend(
[
("cls_token", "vit.embeddings.cls_token"),
("patch_embed.proj.weight", "vit.embeddings.patch_embeddings.projection.weight"),
("patch_embed.proj.bias", "vit.embeddings.patch_embeddings.projection.bias"),
("pos_embed", "vit.embeddings.position_embeddings"),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("norm.weight", "layernorm.weight"),
("norm.bias", "layernorm.bias"),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
lowerCamelCase_ = [(pair[0], pair[1][4:]) if pair[1].startswith("vit" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("norm.weight", "vit.layernorm.weight"),
("norm.bias", "vit.layernorm.bias"),
("head.weight", "classifier.weight"),
("head.bias", "classifier.bias"),
] )
return rename_keys
def __snake_case ( UpperCAmelCase_ : str , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[int]=False ):
for i in range(config.num_hidden_layers ):
if base_model:
lowerCamelCase_ = ""
else:
lowerCamelCase_ = "vit."
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
lowerCamelCase_ = state_dict.pop(F'''blocks.{i}.attn.qkv.weight''' )
lowerCamelCase_ = state_dict.pop(F'''blocks.{i}.attn.qkv.bias''' )
# next, add query, keys and values (in that order) to the state dict
lowerCamelCase_ = in_proj_weight[
: config.hidden_size, :
]
lowerCamelCase_ = in_proj_bias[: config.hidden_size]
lowerCamelCase_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
lowerCamelCase_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
lowerCamelCase_ = in_proj_weight[
-config.hidden_size :, :
]
lowerCamelCase_ = in_proj_bias[-config.hidden_size :]
def __snake_case ( UpperCAmelCase_ : List[Any] ):
lowerCamelCase_ = ["head.weight", "head.bias"]
for k in ignore_keys:
state_dict.pop(UpperCAmelCase_ , UpperCAmelCase_ )
def __snake_case ( UpperCAmelCase_ : Any , UpperCAmelCase_ : str , UpperCAmelCase_ : List[str] ):
lowerCamelCase_ = dct.pop(UpperCAmelCase_ )
lowerCamelCase_ = val
def __snake_case ( ):
lowerCamelCase_ = "http://images.cocodataset.org/val2017/000000039769.jpg"
lowerCamelCase_ = Image.open(requests.get(UpperCAmelCase_ , stream=UpperCAmelCase_ ).raw )
return im
@torch.no_grad()
def __snake_case ( UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Any , UpperCAmelCase_ : List[str]=True ):
lowerCamelCase_ = ViTConfig()
# patch_size
if model_name[-1] == "8":
lowerCamelCase_ = 8
# set labels if required
if not base_model:
lowerCamelCase_ = 1000
lowerCamelCase_ = "huggingface/label-files"
lowerCamelCase_ = "imagenet-1k-id2label.json"
lowerCamelCase_ = json.load(open(hf_hub_download(UpperCAmelCase_ , UpperCAmelCase_ , repo_type="dataset" ) , "r" ) )
lowerCamelCase_ = {int(UpperCAmelCase_ ): v for k, v in idalabel.items()}
lowerCamelCase_ = idalabel
lowerCamelCase_ = {v: k for k, v in idalabel.items()}
# size of the architecture
if model_name in ["dino_vits8", "dino_vits16"]:
lowerCamelCase_ = 384
lowerCamelCase_ = 1536
lowerCamelCase_ = 12
lowerCamelCase_ = 6
# load original model from torch hub
lowerCamelCase_ = torch.hub.load("facebookresearch/dino:main" , UpperCAmelCase_ )
original_model.eval()
# load state_dict of original model, remove and rename some keys
lowerCamelCase_ = original_model.state_dict()
if base_model:
remove_classification_head_(UpperCAmelCase_ )
lowerCamelCase_ = create_rename_keys(UpperCAmelCase_ , base_model=UpperCAmelCase_ )
for src, dest in rename_keys:
rename_key(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
read_in_q_k_v(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
# load HuggingFace model
if base_model:
lowerCamelCase_ = ViTModel(UpperCAmelCase_ , add_pooling_layer=UpperCAmelCase_ ).eval()
else:
lowerCamelCase_ = ViTForImageClassification(UpperCAmelCase_ ).eval()
model.load_state_dict(UpperCAmelCase_ )
# Check outputs on an image, prepared by ViTImageProcessor
lowerCamelCase_ = ViTImageProcessor()
lowerCamelCase_ = image_processor(images=prepare_img() , return_tensors="pt" )
lowerCamelCase_ = encoding["pixel_values"]
lowerCamelCase_ = model(UpperCAmelCase_ )
if base_model:
lowerCamelCase_ = original_model(UpperCAmelCase_ )
assert torch.allclose(UpperCAmelCase_ , outputs.last_hidden_state[:, 0, :] , atol=1E-1 )
else:
lowerCamelCase_ = original_model(UpperCAmelCase_ )
assert logits.shape == outputs.logits.shape
assert torch.allclose(UpperCAmelCase_ , outputs.logits , atol=1E-3 )
Path(UpperCAmelCase_ ).mkdir(exist_ok=UpperCAmelCase_ )
print(F'''Saving model {model_name} to {pytorch_dump_folder_path}''' )
model.save_pretrained(UpperCAmelCase_ )
print(F'''Saving image processor to {pytorch_dump_folder_path}''' )
image_processor.save_pretrained(UpperCAmelCase_ )
if __name__ == "__main__":
a_ : Optional[Any] = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
"""--model_name""",
default="""dino_vitb16""",
type=str,
help="""Name of the model trained with DINO you'd like to convert.""",
)
parser.add_argument(
"""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model directory."""
)
parser.add_argument(
"""--base_model""",
action="""store_true""",
help="""Whether to only convert the base model (no projection head weights).""",
)
parser.set_defaults(base_model=True)
a_ : Optional[Any] = parser.parse_args()
convert_vit_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.base_model)
| 55 |
"""simple docstring"""
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __A (snake_case__):
'''simple docstring'''
@slow
@require_torch
def lowerCAmelCase ( self : Union[str, Any] ) ->Dict:
"""simple docstring"""
snake_case_ = EncoderDecoderModel.from_encoder_decoder_pretrained("""prajjwal1/bert-tiny""" , """prajjwal1/bert-tiny""" )
snake_case_ = BertTokenizer.from_pretrained("""bert-base-uncased""" )
snake_case_ = bertabert.config.encoder.vocab_size
snake_case_ = tokenizer.sep_token_id
snake_case_ = tokenizer.cls_token_id
snake_case_ = 128
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""train[:1%]""" )
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""validation[:1%]""" )
snake_case_ = train_dataset.select(range(32 ) )
snake_case_ = val_dataset.select(range(16 ) )
snake_case_ = 4
def _map_to_encoder_decoder_inputs(UpperCAmelCase_ : int ):
# Tokenizer will automatically set [BOS] <text> [EOS]
snake_case_ = tokenizer(batch["""article"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=512 )
snake_case_ = tokenizer(batch["""highlights"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=128 )
snake_case_ = inputs.input_ids
snake_case_ = inputs.attention_mask
snake_case_ = outputs.input_ids
snake_case_ = outputs.input_ids.copy()
snake_case_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch["""labels"""]
]
snake_case_ = outputs.attention_mask
assert all(len(UpperCAmelCase_ ) == 512 for x in inputs.input_ids )
assert all(len(UpperCAmelCase_ ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(UpperCAmelCase_ : Union[str, Any] ):
snake_case_ = pred.label_ids
snake_case_ = pred.predictions
# all unnecessary tokens are removed
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(UpperCAmelCase_ ) )] ) / len(UpperCAmelCase_ )
return {"accuracy": accuracy}
# map train dataset
snake_case_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
train_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
# same for validation dataset
snake_case_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
val_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
snake_case_ = self.get_auto_remove_tmp_dir()
snake_case_ = SeqaSeqTrainingArguments(
output_dir=UpperCAmelCase_ , per_device_train_batch_size=UpperCAmelCase_ , per_device_eval_batch_size=UpperCAmelCase_ , predict_with_generate=UpperCAmelCase_ , evaluation_strategy="""steps""" , do_train=UpperCAmelCase_ , do_eval=UpperCAmelCase_ , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
snake_case_ = SeqaSeqTrainer(
model=UpperCAmelCase_ , args=UpperCAmelCase_ , compute_metrics=_compute_metrics , train_dataset=UpperCAmelCase_ , eval_dataset=UpperCAmelCase_ , tokenizer=UpperCAmelCase_ , )
# start training
trainer.train()
| 347 | 0 |
'''simple docstring'''
import warnings
warnings.warn(
'memory_utils has been reorganized to utils.memory. Import `find_executable_batchsize` from the main `__init__`: '
'`from accelerate import find_executable_batch_size` to avoid this warning.',
FutureWarning,
)
| 56 |
"""simple docstring"""
# this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.:
# python ./utils/get_modified_files.py utils src tests examples
#
# it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered
# since the output of this script is fed into Makefile commands it doesn't print a newline after the results
import re
import subprocess
import sys
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8')
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split()
__SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:])
__SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""")
__SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)]
print(' '.join(relevant_modified_files), end='')
| 347 | 0 |
"""simple docstring"""
def _lowerCamelCase ( _UpperCamelCase ):
'''simple docstring'''
__lowerCAmelCase = [0] * len(_UpperCamelCase )
for i in range(1 , len(_UpperCamelCase ) ):
# use last results for better performance - dynamic programming
__lowerCAmelCase = prefix_result[i - 1]
while j > 0 and input_string[i] != input_string[j]:
__lowerCAmelCase = prefix_result[j - 1]
if input_string[i] == input_string[j]:
j += 1
__lowerCAmelCase = j
return prefix_result
def _lowerCamelCase ( _UpperCamelCase ):
'''simple docstring'''
return max(prefix_function(_UpperCamelCase ) )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 57 |
"""simple docstring"""
import argparse
import json
import os
import fairseq
import torch
from fairseq.data import Dictionary
from transformers import (
WavaVecaConfig,
WavaVecaCTCTokenizer,
WavaVecaFeatureExtractor,
WavaVecaForCTC,
WavaVecaForPreTraining,
WavaVecaProcessor,
logging,
)
from transformers.models.wavaveca.modeling_wavaveca import WavaVecaForSequenceClassification
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'post_extract_proj': 'feature_projection.projection',
'encoder.pos_conv.0': 'encoder.pos_conv_embed.conv',
'self_attn.k_proj': 'encoder.layers.*.attention.k_proj',
'self_attn.v_proj': 'encoder.layers.*.attention.v_proj',
'self_attn.q_proj': 'encoder.layers.*.attention.q_proj',
'self_attn.out_proj': 'encoder.layers.*.attention.out_proj',
'self_attn_layer_norm': 'encoder.layers.*.layer_norm',
'fc1': 'encoder.layers.*.feed_forward.intermediate_dense',
'fc2': 'encoder.layers.*.feed_forward.output_dense',
'final_layer_norm': 'encoder.layers.*.final_layer_norm',
'encoder.layer_norm': 'encoder.layer_norm',
'adapter_layer': 'encoder.layers.*.adapter_layer',
'w2v_model.layer_norm': 'feature_projection.layer_norm',
'quantizer.weight_proj': 'quantizer.weight_proj',
'quantizer.vars': 'quantizer.codevectors',
'project_q': 'project_q',
'final_proj': 'project_hid',
'w2v_encoder.proj': 'lm_head',
'mask_emb': 'masked_spec_embed',
'pooling_layer.linear': 'projector',
'pooling_layer.projection': 'classifier',
}
__SCREAMING_SNAKE_CASE : List[Any] = [
'lm_head',
'quantizer.weight_proj',
'quantizer.codevectors',
'project_q',
'project_hid',
'projector',
'classifier',
]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = {}
with open(_SCREAMING_SNAKE_CASE , """r""" ) as file:
for line_number, line in enumerate(_SCREAMING_SNAKE_CASE ):
snake_case_ = line.strip()
if line:
snake_case_ = line.split()
snake_case_ = line_number
snake_case_ = words[0]
snake_case_ = value
return result
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
for attribute in key.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ).shape
elif weight_type is not None and weight_type == "param":
snake_case_ = hf_pointer
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = shape_pointer.shape
# let's reduce dimension
snake_case_ = value[0]
else:
snake_case_ = hf_pointer.shape
if hf_shape != value.shape:
raise ValueError(
f"""Shape of hf {key + "." + weight_type if weight_type is not None else ""} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}""" )
if weight_type == "weight":
snake_case_ = value
elif weight_type == "weight_g":
snake_case_ = value
elif weight_type == "weight_v":
snake_case_ = value
elif weight_type == "bias":
snake_case_ = value
elif weight_type == "param":
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = value
else:
snake_case_ = value
logger.info(f"""{key + "." + weight_type if weight_type is not None else ""} was initialized from {full_name}.""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = """.""".join([key, weight_type] )
elif weight_type is not None and weight_type == "param":
snake_case_ = """.""".join([key, hf_param_name] )
else:
snake_case_ = key
snake_case_ = value if """lm_head""" in full_key else value[0]
__SCREAMING_SNAKE_CASE : int = {
'W_a': 'linear_1.weight',
'W_b': 'linear_2.weight',
'b_a': 'linear_1.bias',
'b_b': 'linear_2.bias',
'ln_W': 'norm.weight',
'ln_b': 'norm.bias',
}
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> List[str]:
snake_case_ = False
for key, mapped_key in MAPPING.items():
snake_case_ = """wav2vec2.""" + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key
if key in name or key.split("""w2v_model.""" )[-1] == name.split(""".""" )[0]:
snake_case_ = True
if "*" in mapped_key:
snake_case_ = name.split(_SCREAMING_SNAKE_CASE )[0].split(""".""" )[-2]
snake_case_ = mapped_key.replace("""*""" , _SCREAMING_SNAKE_CASE )
if "weight_g" in name:
snake_case_ = """weight_g"""
elif "weight_v" in name:
snake_case_ = """weight_v"""
elif "bias" in name:
snake_case_ = """bias"""
elif "weight" in name:
# TODO: don't match quantizer.weight_proj
snake_case_ = """weight"""
else:
snake_case_ = None
if hf_dict is not None:
rename_dict(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
else:
set_recursively(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
return is_used
return is_used
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = []
snake_case_ = fairseq_model.state_dict()
snake_case_ = hf_model.wavaveca.feature_extractor
for name, value in fairseq_dict.items():
snake_case_ = False
if "conv_layers" in name:
load_conv_layer(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , hf_model.config.feat_extract_norm == """group""" , )
snake_case_ = True
else:
snake_case_ = load_wavaveca_layer(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not is_used:
unused_weights.append(_SCREAMING_SNAKE_CASE )
logger.warning(f"""Unused weights: {unused_weights}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
snake_case_ = full_name.split("""conv_layers.""" )[-1]
snake_case_ = name.split(""".""" )
snake_case_ = int(items[0] )
snake_case_ = int(items[1] )
if type_id == 0:
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(_SCREAMING_SNAKE_CASE )
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=False ) -> int:
if config_path is not None:
snake_case_ = WavaVecaConfig.from_pretrained(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaConfig()
if is_seq_class:
snake_case_ = read_txt_into_dict(_SCREAMING_SNAKE_CASE )
snake_case_ = idalabel
snake_case_ = WavaVecaForSequenceClassification(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
feature_extractor.save_pretrained(_SCREAMING_SNAKE_CASE )
elif is_finetuned:
if dict_path:
snake_case_ = Dictionary.load(_SCREAMING_SNAKE_CASE )
# important change bos & pad token id since CTC symbol is <pad> and
# not <s> as in fairseq
snake_case_ = target_dict.pad_index
snake_case_ = target_dict.bos_index
snake_case_ = target_dict.eos_index
snake_case_ = len(target_dict.symbols )
snake_case_ = os.path.join(_SCREAMING_SNAKE_CASE , """vocab.json""" )
if not os.path.isdir(_SCREAMING_SNAKE_CASE ):
logger.error("""--pytorch_dump_folder_path ({}) should be a directory""".format(_SCREAMING_SNAKE_CASE ) )
return
os.makedirs(_SCREAMING_SNAKE_CASE , exist_ok=_SCREAMING_SNAKE_CASE )
snake_case_ = target_dict.indices
# fairseq has the <pad> and <s> switched
snake_case_ = 0
snake_case_ = 1
with open(_SCREAMING_SNAKE_CASE , """w""" , encoding="""utf-8""" ) as vocab_handle:
json.dump(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaCTCTokenizer(
_SCREAMING_SNAKE_CASE , unk_token=target_dict.unk_word , pad_token=target_dict.pad_word , bos_token=target_dict.bos_word , eos_token=target_dict.eos_word , word_delimiter_token="""|""" , do_lower_case=_SCREAMING_SNAKE_CASE , )
snake_case_ = True if config.feat_extract_norm == """layer""" else False
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
snake_case_ = WavaVecaProcessor(feature_extractor=_SCREAMING_SNAKE_CASE , tokenizer=_SCREAMING_SNAKE_CASE )
processor.save_pretrained(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaForCTC(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaForPreTraining(_SCREAMING_SNAKE_CASE )
if is_finetuned or is_seq_class:
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={"""data""": """/""".join(dict_path.split("""/""" )[:-1] )} )
else:
snake_case_ = argparse.Namespace(task="""audio_pretraining""" )
snake_case_ = fairseq.tasks.setup_task(_SCREAMING_SNAKE_CASE )
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] , task=_SCREAMING_SNAKE_CASE )
snake_case_ = model[0].eval()
recursively_load_weights(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , not is_finetuned )
hf_wavavec.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : str = argparse.ArgumentParser()
parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.')
parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to fairseq checkpoint')
parser.add_argument('--dict_path', default=None, type=str, help='Path to dict of fine-tuned model')
parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert')
parser.add_argument(
'--not_finetuned', action='store_true', help='Whether the model to convert is a fine-tuned model or not'
)
parser.add_argument(
'--is_seq_class',
action='store_true',
help='Whether the model to convert is a fine-tuned sequence classification model or not',
)
__SCREAMING_SNAKE_CASE : Any = parser.parse_args()
__SCREAMING_SNAKE_CASE : List[Any] = not args.not_finetuned and not args.is_seq_class
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.config_path,
args.dict_path,
is_finetuned,
args.is_seq_class,
)
| 347 | 0 |
'''simple docstring'''
import unittest
from transformers import (
MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING,
TF_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING,
TextClassificationPipeline,
pipeline,
)
from transformers.testing_utils import is_pipeline_test, nested_simplify, require_tf, require_torch, slow
from .test_pipelines_common import ANY
# These 2 model types require different inputs than those of the usual text models.
lowercase_ = {"""LayoutLMv2Config""", """LayoutLMv3Config"""}
@is_pipeline_test
class a_ ( unittest.TestCase ):
'''simple docstring'''
UpperCamelCase = MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING
UpperCamelCase = TF_MODEL_FOR_SEQUENCE_CLASSIFICATION_MAPPING
if model_mapping is not None:
UpperCamelCase = {config: model for config, model in model_mapping.items() if config.__name__ not in _TO_SKIP}
if tf_model_mapping is not None:
UpperCamelCase = {
config: model for config, model in tf_model_mapping.items() if config.__name__ not in _TO_SKIP
}
@require_torch
def snake_case_( self ) -> Union[str, Any]:
_SCREAMING_SNAKE_CASE = pipeline(
task="""text-classification""" , model="""hf-internal-testing/tiny-random-distilbert""" , framework="""pt""" )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}] )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" , top_k=2 )
self.assertEqual(
nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}] )
_SCREAMING_SNAKE_CASE = text_classifier(["""This is great !""", """This is bad"""] , top_k=2 )
self.assertEqual(
nested_simplify(A ) , [
[{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}],
[{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}],
] , )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" , top_k=1 )
self.assertEqual(nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}] )
# Legacy behavior
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" , return_all_scores=A )
self.assertEqual(nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}] )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" , return_all_scores=A )
self.assertEqual(
nested_simplify(A ) , [[{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}]] )
_SCREAMING_SNAKE_CASE = text_classifier(["""This is great !""", """Something else"""] , return_all_scores=A )
self.assertEqual(
nested_simplify(A ) , [
[{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}],
[{"""label""": """LABEL_0""", """score""": 0.504}, {"""label""": """LABEL_1""", """score""": 0.496}],
] , )
_SCREAMING_SNAKE_CASE = text_classifier(["""This is great !""", """Something else"""] , return_all_scores=A )
self.assertEqual(
nested_simplify(A ) , [
{"""label""": """LABEL_0""", """score""": 0.504},
{"""label""": """LABEL_0""", """score""": 0.504},
] , )
@require_torch
def snake_case_( self ) -> int:
import torch
_SCREAMING_SNAKE_CASE = pipeline(
task="""text-classification""" , model="""hf-internal-testing/tiny-random-distilbert""" , framework="""pt""" , device=torch.device("""cpu""" ) , )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}] )
@require_tf
def snake_case_( self ) -> List[Any]:
_SCREAMING_SNAKE_CASE = pipeline(
task="""text-classification""" , model="""hf-internal-testing/tiny-random-distilbert""" , framework="""tf""" )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """LABEL_0""", """score""": 0.504}] )
@slow
@require_torch
def snake_case_( self ) -> Any:
_SCREAMING_SNAKE_CASE = pipeline("""text-classification""" )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """POSITIVE""", """score""": 1.0}] )
_SCREAMING_SNAKE_CASE = text_classifier("""This is bad !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """NEGATIVE""", """score""": 1.0}] )
_SCREAMING_SNAKE_CASE = text_classifier("""Birds are a type of animal""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """POSITIVE""", """score""": 0.988}] )
@slow
@require_tf
def snake_case_( self ) -> List[Any]:
_SCREAMING_SNAKE_CASE = pipeline("""text-classification""" , framework="""tf""" )
_SCREAMING_SNAKE_CASE = text_classifier("""This is great !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """POSITIVE""", """score""": 1.0}] )
_SCREAMING_SNAKE_CASE = text_classifier("""This is bad !""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """NEGATIVE""", """score""": 1.0}] )
_SCREAMING_SNAKE_CASE = text_classifier("""Birds are a type of animal""" )
self.assertEqual(nested_simplify(A ) , [{"""label""": """POSITIVE""", """score""": 0.988}] )
def snake_case_( self , A , A , A ) -> Optional[Any]:
_SCREAMING_SNAKE_CASE = TextClassificationPipeline(model=A , tokenizer=A )
return text_classifier, ["HuggingFace is in", "This is another test"]
def snake_case_( self , A , A ) -> int:
_SCREAMING_SNAKE_CASE = text_classifier.model
# Small inputs because BartTokenizer tiny has maximum position embeddings = 22
_SCREAMING_SNAKE_CASE = """HuggingFace is in"""
_SCREAMING_SNAKE_CASE = text_classifier(A )
self.assertEqual(nested_simplify(A ) , [{"""label""": ANY(A ), """score""": ANY(A )}] )
self.assertTrue(outputs[0]["""label"""] in model.config.idalabel.values() )
_SCREAMING_SNAKE_CASE = ["""HuggingFace is in """, """Paris is in France"""]
_SCREAMING_SNAKE_CASE = text_classifier(A )
self.assertEqual(
nested_simplify(A ) , [{"""label""": ANY(A ), """score""": ANY(A )}, {"""label""": ANY(A ), """score""": ANY(A )}] , )
self.assertTrue(outputs[0]["""label"""] in model.config.idalabel.values() )
self.assertTrue(outputs[1]["""label"""] in model.config.idalabel.values() )
# Forcing to get all results with `top_k=None`
# This is NOT the legacy format
_SCREAMING_SNAKE_CASE = text_classifier(A , top_k=A )
_SCREAMING_SNAKE_CASE = len(model.config.idalabel.values() )
self.assertEqual(
nested_simplify(A ) , [[{"""label""": ANY(A ), """score""": ANY(A )}] * N, [{"""label""": ANY(A ), """score""": ANY(A )}] * N] , )
_SCREAMING_SNAKE_CASE = {"""text""": """HuggingFace is in """, """text_pair""": """Paris is in France"""}
_SCREAMING_SNAKE_CASE = text_classifier(A )
self.assertEqual(
nested_simplify(A ) , {"""label""": ANY(A ), """score""": ANY(A )} , )
self.assertTrue(outputs["""label"""] in model.config.idalabel.values() )
# This might be used a text pair, but tokenizer + pipe interaction
# makes it hard to understand that it's not using the pair properly
# https://github.com/huggingface/transformers/issues/17305
# We disabled this usage instead as it was outputting wrong outputs.
_SCREAMING_SNAKE_CASE = [["""HuggingFace is in """, """Paris is in France"""]]
with self.assertRaises(A ):
text_classifier(A )
# This used to be valid for doing text pairs
# We're keeping it working because of backward compatibility
_SCREAMING_SNAKE_CASE = text_classifier([[["""HuggingFace is in """, """Paris is in France"""]]] )
self.assertEqual(
nested_simplify(A ) , [{"""label""": ANY(A ), """score""": ANY(A )}] , )
self.assertTrue(outputs[0]["""label"""] in model.config.idalabel.values() )
| 58 |
"""simple docstring"""
import tempfile
import unittest
import numpy as np
import transformers
from transformers import GPTaTokenizer, GPTJConfig, is_flax_available, is_torch_available
from transformers.testing_utils import is_pt_flax_cross_test, require_flax, tooslow
from ...generation.test_flax_utils import FlaxGenerationTesterMixin
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.modeling_flax_pytorch_utils import (
convert_pytorch_state_dict_to_flax,
load_flax_weights_in_pytorch_model,
)
from transformers.models.gptj.modeling_flax_gptj import FlaxGPTJForCausalLM, FlaxGPTJModel
if is_torch_available():
import torch
class __A :
'''simple docstring'''
def __init__( self : Dict , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Any=14 , UpperCAmelCase_ : Union[str, Any]=7 , UpperCAmelCase_ : Tuple=True , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Union[str, Any]=False , UpperCAmelCase_ : Union[str, Any]=True , UpperCAmelCase_ : str=99 , UpperCAmelCase_ : Union[str, Any]=32 , UpperCAmelCase_ : List[Any]=4 , UpperCAmelCase_ : Optional[int]=4 , UpperCAmelCase_ : int=4 , UpperCAmelCase_ : str=37 , UpperCAmelCase_ : Any="gelu" , UpperCAmelCase_ : str=0.1 , UpperCAmelCase_ : Union[str, Any]=0.1 , UpperCAmelCase_ : int=512 , UpperCAmelCase_ : Tuple=0.02 , ) ->List[str]:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_input_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = rotary_dim
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = initializer_range
snake_case_ = None
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
def lowerCAmelCase ( self : int ) ->Optional[int]:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_input_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = GPTJConfig(
vocab_size=self.vocab_size , n_embd=self.hidden_size , n_layer=self.num_hidden_layers , n_head=self.num_attention_heads , n_positions=self.max_position_embeddings , use_cache=UpperCAmelCase_ , bos_token_id=self.bos_token_id , eos_token_id=self.eos_token_id , pad_token_id=self.pad_token_id , rotary_dim=self.rotary_dim , )
return (config, input_ids, input_mask)
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """attention_mask""": attention_mask}
return config, inputs_dict
def lowerCAmelCase ( self : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.ones((input_ids.shape[0], max_decoder_length) , dtype="""i4""" )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , attention_mask=UpperCAmelCase_ , past_key_values=outputs_cache.past_key_values , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
def lowerCAmelCase ( self : Union[str, Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = jnp.concatenate(
[attention_mask, jnp.zeros((attention_mask.shape[0], max_decoder_length - attention_mask.shape[1]) )] , axis=-1 , )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , past_key_values=outputs_cache.past_key_values , attention_mask=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
@require_flax
class __A (snake_case__ , snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Any = (FlaxGPTJModel, FlaxGPTJForCausalLM) if is_flax_available() else ()
__lowercase: List[str] = (FlaxGPTJForCausalLM,) if is_flax_available() else ()
def lowerCAmelCase ( self : Tuple ) ->List[str]:
"""simple docstring"""
snake_case_ = FlaxGPTJModelTester(self )
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] ) ->Any:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward_with_attn_mask(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
@tooslow
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = GPTaTokenizer.from_pretrained("""gpt2""" , pad_token="""<|endoftext|>""" , padding_side="""left""" )
snake_case_ = tokenizer(["""Hello this is a long string""", """Hey"""] , return_tensors="""np""" , padding=UpperCAmelCase_ , truncation=UpperCAmelCase_ )
snake_case_ = FlaxGPTJForCausalLM.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = False
snake_case_ = model.config.eos_token_id
snake_case_ = jax.jit(model.generate )
snake_case_ = jit_generate(
inputs["""input_ids"""] , attention_mask=inputs["""attention_mask"""] , pad_token_id=tokenizer.pad_token_id ).sequences
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = [
"""Hello this is a long string of text.\n\nI'm trying to get the text of the""",
"""Hey, I'm a little late to the party. I'm going to""",
]
self.assertListEqual(UpperCAmelCase_ , UpperCAmelCase_ )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = convert_pytorch_state_dict_to_flax(pt_model.state_dict() , UpperCAmelCase_ )
snake_case_ = fx_state
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
pt_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = model_class.from_pretrained(UpperCAmelCase_ , from_pt=UpperCAmelCase_ )
snake_case_ = fx_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output_loaded, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output_loaded[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = load_flax_weights_in_pytorch_model(UpperCAmelCase_ , fx_model.params )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
# make sure weights are tied in PyTorch
pt_model.tie_weights()
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
fx_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = pt_model_class.from_pretrained(UpperCAmelCase_ , from_flax=UpperCAmelCase_ )
with torch.no_grad():
snake_case_ = pt_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@tooslow
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
import collections
import inspect
import unittest
from typing import Dict, List, Tuple
from transformers import MaskFormerSwinConfig
from transformers.testing_utils import require_torch, require_torch_multi_gpu, torch_device
from transformers.utils import is_torch_available
from ...test_backbone_common import BackboneTesterMixin
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import MaskFormerSwinBackbone
from transformers.models.maskformer import MaskFormerSwinModel
class UpperCAmelCase :
def __init__(self : Dict , snake_case__ : Dict , snake_case__ : Any=13 , snake_case__ : Any=32 , snake_case__ : Optional[Any]=2 , snake_case__ : Union[str, Any]=3 , snake_case__ : List[Any]=16 , snake_case__ : int=[1, 2, 1] , snake_case__ : Dict=[2, 2, 4] , snake_case__ : Dict=2 , snake_case__ : Tuple=2.0 , snake_case__ : Optional[int]=True , snake_case__ : Union[str, Any]=0.0 , snake_case__ : Any=0.0 , snake_case__ : Union[str, Any]=0.1 , snake_case__ : int="gelu" , snake_case__ : Optional[int]=False , snake_case__ : List[Any]=True , snake_case__ : List[str]=0.02 , snake_case__ : int=1e-5 , snake_case__ : List[str]=True , snake_case__ : Union[str, Any]=None , snake_case__ : List[Any]=True , snake_case__ : Optional[Any]=10 , snake_case__ : Optional[Any]=8 , snake_case__ : Any=["stage1", "stage2", "stage3"] , snake_case__ : Tuple=[1, 2, 3] , ) -> Union[str, Any]:
'''simple docstring'''
snake_case : Any = parent
snake_case : Optional[int] = batch_size
snake_case : Union[str, Any] = image_size
snake_case : Dict = patch_size
snake_case : Optional[Any] = num_channels
snake_case : Union[str, Any] = embed_dim
snake_case : int = depths
snake_case : List[str] = num_heads
snake_case : Union[str, Any] = window_size
snake_case : Union[str, Any] = mlp_ratio
snake_case : List[Any] = qkv_bias
snake_case : List[Any] = hidden_dropout_prob
snake_case : Union[str, Any] = attention_probs_dropout_prob
snake_case : Union[str, Any] = drop_path_rate
snake_case : int = hidden_act
snake_case : Optional[int] = use_absolute_embeddings
snake_case : int = patch_norm
snake_case : Union[str, Any] = layer_norm_eps
snake_case : Any = initializer_range
snake_case : Optional[Any] = is_training
snake_case : Tuple = scope
snake_case : Optional[int] = use_labels
snake_case : Optional[Any] = type_sequence_label_size
snake_case : Union[str, Any] = encoder_stride
snake_case : Any = out_features
snake_case : Tuple = out_indices
def _SCREAMING_SNAKE_CASE (self : Optional[Any] ) -> Dict:
'''simple docstring'''
snake_case : Optional[Any] = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
snake_case : int = None
if self.use_labels:
snake_case : Tuple = ids_tensor([self.batch_size] , self.type_sequence_label_size )
snake_case : Dict = self.get_config()
return config, pixel_values, labels
def _SCREAMING_SNAKE_CASE (self : List[str] ) -> int:
'''simple docstring'''
return MaskFormerSwinConfig(
image_size=self.image_size , patch_size=self.patch_size , num_channels=self.num_channels , embed_dim=self.embed_dim , depths=self.depths , num_heads=self.num_heads , window_size=self.window_size , mlp_ratio=self.mlp_ratio , qkv_bias=self.qkv_bias , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , drop_path_rate=self.drop_path_rate , hidden_act=self.hidden_act , use_absolute_embeddings=self.use_absolute_embeddings , path_norm=self.patch_norm , layer_norm_eps=self.layer_norm_eps , initializer_range=self.initializer_range , encoder_stride=self.encoder_stride , out_features=self.out_features , out_indices=self.out_indices , )
def _SCREAMING_SNAKE_CASE (self : Tuple , snake_case__ : List[Any] , snake_case__ : List[str] , snake_case__ : Tuple ) -> Optional[Any]:
'''simple docstring'''
snake_case : Union[str, Any] = MaskFormerSwinModel(config=snake_case__ )
model.to(snake_case__ )
model.eval()
snake_case : List[Any] = model(snake_case__ )
snake_case : Dict = ((config.image_size // config.patch_size) ** 2) // (4 ** (len(config.depths ) - 1))
snake_case : int = int(config.embed_dim * 2 ** (len(config.depths ) - 1) )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, expected_seq_len, expected_dim) )
def _SCREAMING_SNAKE_CASE (self : List[str] , snake_case__ : Union[str, Any] , snake_case__ : List[Any] , snake_case__ : Union[str, Any] ) -> str:
'''simple docstring'''
snake_case : Optional[int] = MaskFormerSwinBackbone(config=snake_case__ )
model.to(snake_case__ )
model.eval()
snake_case : List[Any] = model(snake_case__ )
# verify feature maps
self.parent.assertEqual(len(result.feature_maps ) , len(config.out_features ) )
self.parent.assertListEqual(list(result.feature_maps[0].shape ) , [13, 16, 16, 16] )
# verify channels
self.parent.assertEqual(len(model.channels ) , len(config.out_features ) )
self.parent.assertListEqual(model.channels , [16, 32, 64] )
# verify ValueError
with self.parent.assertRaises(snake_case__ ):
snake_case : Tuple = ["stem"]
snake_case : List[Any] = MaskFormerSwinBackbone(config=snake_case__ )
def _SCREAMING_SNAKE_CASE (self : List[Any] ) -> List[Any]:
'''simple docstring'''
snake_case : Union[str, Any] = self.prepare_config_and_inputs()
snake_case , snake_case , snake_case : List[Any] = config_and_inputs
snake_case : int = {"pixel_values": pixel_values}
return config, inputs_dict
@require_torch
class UpperCAmelCase ( A_ ,A_ ,unittest.TestCase ):
A__ : List[str] = (
(
MaskFormerSwinModel,
MaskFormerSwinBackbone,
)
if is_torch_available()
else ()
)
A__ : str = {"feature-extraction": MaskFormerSwinModel} if is_torch_available() else {}
A__ : Optional[Any] = False
A__ : List[Any] = False
A__ : List[str] = False
A__ : List[str] = False
A__ : Union[str, Any] = False
def _SCREAMING_SNAKE_CASE (self : Optional[int] ) -> List[str]:
'''simple docstring'''
snake_case : str = MaskFormerSwinModelTester(self )
snake_case : Optional[int] = ConfigTester(self , config_class=snake_case__ , embed_dim=37 )
@require_torch_multi_gpu
@unittest.skip(
reason=(
"`MaskFormerSwinModel` outputs `hidden_states_spatial_dimensions` which doesn't work well with"
" `nn.DataParallel`"
) )
def _SCREAMING_SNAKE_CASE (self : str ) -> Optional[Any]:
'''simple docstring'''
pass
def _SCREAMING_SNAKE_CASE (self : str ) -> List[str]:
'''simple docstring'''
self.create_and_test_config_common_properties()
self.config_tester.create_and_test_config_to_json_string()
self.config_tester.create_and_test_config_to_json_file()
self.config_tester.create_and_test_config_from_and_save_pretrained()
self.config_tester.create_and_test_config_with_num_labels()
self.config_tester.check_config_can_be_init_without_params()
self.config_tester.check_config_arguments_init()
def _SCREAMING_SNAKE_CASE (self : Tuple ) -> List[Any]:
'''simple docstring'''
return
def _SCREAMING_SNAKE_CASE (self : Dict ) -> str:
'''simple docstring'''
snake_case : str = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*snake_case__ )
def _SCREAMING_SNAKE_CASE (self : int ) -> Dict:
'''simple docstring'''
snake_case : Any = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_backbone(*snake_case__ )
@unittest.skip("Swin does not use inputs_embeds" )
def _SCREAMING_SNAKE_CASE (self : int ) -> Any:
'''simple docstring'''
pass
@unittest.skip("Swin does not support feedforward chunking" )
def _SCREAMING_SNAKE_CASE (self : List[str] ) -> Dict:
'''simple docstring'''
pass
def _SCREAMING_SNAKE_CASE (self : Optional[Any] ) -> List[str]:
'''simple docstring'''
snake_case , snake_case : Any = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case : int = model_class(snake_case__ )
self.assertIsInstance(model.get_input_embeddings() , (nn.Module) )
snake_case : List[Any] = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(snake_case__ , nn.Linear ) )
def _SCREAMING_SNAKE_CASE (self : Optional[Any] ) -> Dict:
'''simple docstring'''
snake_case , snake_case : str = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
snake_case : str = model_class(snake_case__ )
snake_case : Optional[int] = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
snake_case : Optional[Any] = [*signature.parameters.keys()]
snake_case : Tuple = ["pixel_values"]
self.assertListEqual(arg_names[:1] , snake_case__ )
@unittest.skip(reason="MaskFormerSwin is only used as backbone and doesn't support output_attentions" )
def _SCREAMING_SNAKE_CASE (self : List[Any] ) -> str:
'''simple docstring'''
pass
@unittest.skip(reason="MaskFormerSwin is only used as an internal backbone" )
def _SCREAMING_SNAKE_CASE (self : Tuple ) -> Any:
'''simple docstring'''
pass
def _SCREAMING_SNAKE_CASE (self : Optional[Any] , snake_case__ : List[Any] , snake_case__ : str , snake_case__ : List[Any] , snake_case__ : Tuple ) -> Optional[int]:
'''simple docstring'''
snake_case : Tuple = model_class(snake_case__ )
model.to(snake_case__ )
model.eval()
with torch.no_grad():
snake_case : Any = model(**self._prepare_for_class(snake_case__ , snake_case__ ) )
snake_case : int = outputs.hidden_states
snake_case : Union[str, Any] = getattr(
self.model_tester , "expected_num_hidden_layers" , len(self.model_tester.depths ) + 1 )
self.assertEqual(len(snake_case__ ) , snake_case__ )
# Swin has a different seq_length
snake_case : Any = (
config.patch_size
if isinstance(config.patch_size , collections.abc.Iterable )
else (config.patch_size, config.patch_size)
)
snake_case : Tuple = (image_size[1] // patch_size[1]) * (image_size[0] // patch_size[0])
self.assertListEqual(
list(hidden_states[0].shape[-2:] ) , [num_patches, self.model_tester.embed_dim] , )
def _SCREAMING_SNAKE_CASE (self : Dict ) -> Union[str, Any]:
'''simple docstring'''
snake_case , snake_case : List[Any] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case : int = (
self.model_tester.image_size
if isinstance(self.model_tester.image_size , collections.abc.Iterable )
else (self.model_tester.image_size, self.model_tester.image_size)
)
for model_class in self.all_model_classes:
snake_case : int = True
self.check_hidden_states_output(snake_case__ , snake_case__ , snake_case__ , snake_case__ )
# check that output_hidden_states also work using config
del inputs_dict["output_hidden_states"]
snake_case : Dict = True
self.check_hidden_states_output(snake_case__ , snake_case__ , snake_case__ , snake_case__ )
def _SCREAMING_SNAKE_CASE (self : int ) -> Any:
'''simple docstring'''
snake_case , snake_case : Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
snake_case : Any = 3
snake_case : List[str] = (
self.model_tester.image_size
if isinstance(self.model_tester.image_size , collections.abc.Iterable )
else (self.model_tester.image_size, self.model_tester.image_size)
)
snake_case : Tuple = (
config.patch_size
if isinstance(config.patch_size , collections.abc.Iterable )
else (config.patch_size, config.patch_size)
)
snake_case : str = image_size[0] + patch_size[0] - (image_size[0] % patch_size[0])
snake_case : str = image_size[1] + patch_size[1] - (image_size[1] % patch_size[1])
for model_class in self.all_model_classes:
snake_case : str = True
self.check_hidden_states_output(snake_case__ , snake_case__ , snake_case__ , (padded_height, padded_width) )
# check that output_hidden_states also work using config
del inputs_dict["output_hidden_states"]
snake_case : Optional[Any] = True
self.check_hidden_states_output(snake_case__ , snake_case__ , snake_case__ , (padded_height, padded_width) )
@unittest.skip(reason="MaskFormerSwin doesn't have pretrained checkpoints" )
def _SCREAMING_SNAKE_CASE (self : Optional[int] ) -> str:
'''simple docstring'''
pass
@unittest.skip(reason="This will be fixed once MaskFormerSwin is replaced by native Swin" )
def _SCREAMING_SNAKE_CASE (self : str ) -> int:
'''simple docstring'''
pass
@unittest.skip(reason="This will be fixed once MaskFormerSwin is replaced by native Swin" )
def _SCREAMING_SNAKE_CASE (self : int ) -> str:
'''simple docstring'''
pass
def _SCREAMING_SNAKE_CASE (self : Any ) -> Any:
'''simple docstring'''
snake_case , snake_case : List[Any] = self.model_tester.prepare_config_and_inputs_for_common()
def set_nan_tensor_to_zero(snake_case__ : Union[str, Any] ):
snake_case : Any = 0
return t
def check_equivalence(snake_case__ : Union[str, Any] , snake_case__ : int , snake_case__ : List[str] , snake_case__ : Optional[int]={} ):
with torch.no_grad():
snake_case : Optional[Any] = model(**snake_case__ , return_dict=snake_case__ , **snake_case__ )
snake_case : Tuple = model(**snake_case__ , return_dict=snake_case__ , **snake_case__ ).to_tuple()
def recursive_check(snake_case__ : List[str] , snake_case__ : Optional[Any] ):
if isinstance(snake_case__ , (List, Tuple) ):
for tuple_iterable_value, dict_iterable_value in zip(snake_case__ , snake_case__ ):
recursive_check(snake_case__ , snake_case__ )
elif isinstance(snake_case__ , snake_case__ ):
for tuple_iterable_value, dict_iterable_value in zip(
tuple_object.values() , dict_object.values() ):
recursive_check(snake_case__ , snake_case__ )
elif tuple_object is None:
return
else:
self.assertTrue(
torch.allclose(
set_nan_tensor_to_zero(snake_case__ ) , set_nan_tensor_to_zero(snake_case__ ) , atol=1e-5 ) , msg=(
"Tuple and dict output are not equal. Difference:"
f""" {torch.max(torch.abs(tuple_object - dict_object ) )}. Tuple has `nan`:"""
f""" {torch.isnan(snake_case__ ).any()} and `inf`: {torch.isinf(snake_case__ )}. Dict has"""
f""" `nan`: {torch.isnan(snake_case__ ).any()} and `inf`: {torch.isinf(snake_case__ )}."""
) , )
recursive_check(snake_case__ , snake_case__ )
for model_class in self.all_model_classes:
snake_case : Optional[int] = model_class(snake_case__ )
model.to(snake_case__ )
model.eval()
snake_case : Union[str, Any] = self._prepare_for_class(snake_case__ , snake_case__ )
snake_case : Tuple = self._prepare_for_class(snake_case__ , snake_case__ )
check_equivalence(snake_case__ , snake_case__ , snake_case__ )
snake_case : Tuple = self._prepare_for_class(snake_case__ , snake_case__ , return_labels=snake_case__ )
snake_case : Optional[Any] = self._prepare_for_class(snake_case__ , snake_case__ , return_labels=snake_case__ )
check_equivalence(snake_case__ , snake_case__ , snake_case__ )
snake_case : Dict = self._prepare_for_class(snake_case__ , snake_case__ )
snake_case : List[Any] = self._prepare_for_class(snake_case__ , snake_case__ )
check_equivalence(snake_case__ , snake_case__ , snake_case__ , {"output_hidden_states": True} )
snake_case : Any = self._prepare_for_class(snake_case__ , snake_case__ , return_labels=snake_case__ )
snake_case : List[str] = self._prepare_for_class(snake_case__ , snake_case__ , return_labels=snake_case__ )
check_equivalence(snake_case__ , snake_case__ , snake_case__ , {"output_hidden_states": True} )
@require_torch
class UpperCAmelCase ( unittest.TestCase ,A_ ):
A__ : int = (MaskFormerSwinBackbone,) if is_torch_available() else ()
A__ : int = MaskFormerSwinConfig
def _SCREAMING_SNAKE_CASE (self : List[Any] ) -> Any:
'''simple docstring'''
snake_case : Union[str, Any] = MaskFormerSwinModelTester(self )
def _SCREAMING_SNAKE_CASE (self : Optional[int] ) -> Optional[Any]:
'''simple docstring'''
snake_case , snake_case : Dict = self.model_tester.prepare_config_and_inputs_for_common()
snake_case : Optional[int] = inputs_dict["pixel_values"].shape[0]
for backbone_class in self.all_model_classes:
snake_case : Optional[int] = backbone_class(snake_case__ )
backbone.to(snake_case__ )
backbone.eval()
snake_case : Union[str, Any] = backbone(**snake_case__ )
# Test default outputs and verify feature maps
self.assertIsInstance(outputs.feature_maps , snake_case__ )
self.assertTrue(len(outputs.feature_maps ) == len(backbone.channels ) )
for feature_map, n_channels in zip(outputs.feature_maps , backbone.channels ):
self.assertTrue(feature_map.shape[:2] , (batch_size, n_channels) )
self.assertIsNone(outputs.hidden_states )
self.assertIsNone(outputs.attentions )
# Test output_hidden_states=True
snake_case : Optional[int] = backbone(**snake_case__ , output_hidden_states=snake_case__ )
self.assertIsNotNone(outputs.hidden_states )
self.assertTrue(len(outputs.hidden_states ) , len(backbone.stage_names ) )
# We skip the stem layer
for hidden_states, n_channels in zip(outputs.hidden_states[1:] , backbone.channels ):
for hidden_state in hidden_states:
# Hidden states are in the format (batch_size, (height * width), n_channels)
snake_case , snake_case , snake_case : Dict = hidden_state.shape
self.assertTrue((h_batch_size, h_n_channels) , (batch_size, n_channels) )
# Test output_attentions=True
if self.has_attentions:
snake_case : Optional[Any] = backbone(**snake_case__ , output_attentions=snake_case__ )
self.assertIsNotNone(outputs.attentions )
| 59 |
"""simple docstring"""
import copy
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ..auto.configuration_auto import CONFIG_MAPPING
__SCREAMING_SNAKE_CASE : Any = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """upernet"""
def __init__( self : str , UpperCAmelCase_ : List[str]=None , UpperCAmelCase_ : str=512 , UpperCAmelCase_ : int=0.02 , UpperCAmelCase_ : Optional[Any]=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Tuple=0.4 , UpperCAmelCase_ : Tuple=384 , UpperCAmelCase_ : Union[str, Any]=256 , UpperCAmelCase_ : str=1 , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : Dict , ) ->Union[str, Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
if backbone_config is None:
logger.info("""`backbone_config` is `None`. Initializing the config with the default `ResNet` backbone.""" )
snake_case_ = CONFIG_MAPPING["""resnet"""](out_features=["""stage1""", """stage2""", """stage3""", """stage4"""] )
elif isinstance(UpperCAmelCase_ , UpperCAmelCase_ ):
snake_case_ = backbone_config.get("""model_type""" )
snake_case_ = CONFIG_MAPPING[backbone_model_type]
snake_case_ = config_class.from_dict(UpperCAmelCase_ )
snake_case_ = backbone_config
snake_case_ = hidden_size
snake_case_ = initializer_range
snake_case_ = pool_scales
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_in_channels
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = loss_ignore_index
def lowerCAmelCase ( self : str ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = copy.deepcopy(self.__dict__ )
snake_case_ = self.backbone_config.to_dict()
snake_case_ = self.__class__.model_type
return output
| 347 | 0 |
"""simple docstring"""
def _snake_case ( _snake_case : list ):
for i in range(len(_snake_case ) - 1 , 0 , -1 ):
lowerCAmelCase : int = False
for j in range(_snake_case , 0 , -1 ):
if unsorted[j] < unsorted[j - 1]:
lowerCAmelCase, lowerCAmelCase : Tuple = unsorted[j - 1], unsorted[j]
lowerCAmelCase : Optional[Any] = True
for j in range(_snake_case ):
if unsorted[j] > unsorted[j + 1]:
lowerCAmelCase, lowerCAmelCase : Any = unsorted[j + 1], unsorted[j]
lowerCAmelCase : int = True
if not swapped:
break
return unsorted
if __name__ == "__main__":
import doctest
doctest.testmod()
snake_case__ : List[str] = input('''Enter numbers separated by a comma:\n''').strip()
snake_case__ : str = [int(item) for item in user_input.split(''',''')]
print(f"""{cocktail_shaker_sort(unsorted) = }""")
| 60 |
"""simple docstring"""
import os
import shutil
import tempfile
import unittest
import numpy as np
from transformers import AutoTokenizer, BarkProcessor
from transformers.testing_utils import require_torch, slow
@require_torch
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = """ylacombe/bark-small"""
snake_case_ = tempfile.mkdtemp()
snake_case_ = """en_speaker_1"""
snake_case_ = """This is a test string"""
snake_case_ = """speaker_embeddings_path.json"""
snake_case_ = """speaker_embeddings"""
def lowerCAmelCase ( self : List[str] , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
return AutoTokenizer.from_pretrained(self.checkpoint , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
shutil.rmtree(self.tmpdirname )
def lowerCAmelCase ( self : List[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
snake_case_ = BarkProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
@slow
def lowerCAmelCase ( self : Dict ) ->int:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
processor.save_pretrained(
self.tmpdirname , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , speaker_embeddings_directory=self.speaker_embeddings_directory , )
snake_case_ = self.get_tokenizer(bos_token="""(BOS)""" , eos_token="""(EOS)""" )
snake_case_ = BarkProcessor.from_pretrained(
self.tmpdirname , self.speaker_embeddings_dict_path , bos_token="""(BOS)""" , eos_token="""(EOS)""" , )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
def lowerCAmelCase ( self : Optional[Any] ) ->Any:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
snake_case_ = 35
snake_case_ = 2
snake_case_ = 8
snake_case_ = {
"""semantic_prompt""": np.ones(UpperCAmelCase_ ),
"""coarse_prompt""": np.ones((nb_codebooks_coarse, seq_len) ),
"""fine_prompt""": np.ones((nb_codebooks_total, seq_len) ),
}
# test providing already loaded voice_preset
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from npz file
snake_case_ = os.path.join(self.tmpdirname , """file.npz""" )
np.savez(UpperCAmelCase_ , **UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from the hub
snake_case_ = processor(text=self.input_string , voice_preset=self.voice_preset )
def lowerCAmelCase ( self : Tuple ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string )
snake_case_ = tokenizer(
self.input_string , padding="""max_length""" , max_length=256 , add_special_tokens=UpperCAmelCase_ , return_attention_mask=UpperCAmelCase_ , return_token_type_ids=UpperCAmelCase_ , )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key].squeeze().tolist() )
| 347 | 0 |
"""simple docstring"""
import inspect
import unittest
from huggingface_hub import hf_hub_download
from transformers import ASTConfig
from transformers.testing_utils import require_torch, require_torchaudio, slow, torch_device
from transformers.utils import cached_property, is_torch_available, is_torchaudio_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import ASTForAudioClassification, ASTModel
from transformers.models.audio_spectrogram_transformer.modeling_audio_spectrogram_transformer import (
AUDIO_SPECTROGRAM_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
)
if is_torchaudio_available():
import torchaudio
from transformers import ASTFeatureExtractor
class A_ :
'''simple docstring'''
def __init__( self , lowercase_ , lowercase_=13 , lowercase_=2 , lowercase_=24 , lowercase_=16 , lowercase_=True , lowercase_=True , lowercase_=32 , lowercase_=5 , lowercase_=4 , lowercase_=37 , lowercase_="gelu" , lowercase_=0.1 , lowercase_=0.1 , lowercase_=10 , lowercase_=0.02 , lowercase_=None , lowercase_=2 , lowercase_=2 , ):
"""simple docstring"""
UpperCAmelCase_ : Optional[int] = parent
UpperCAmelCase_ : Union[str, Any] = batch_size
UpperCAmelCase_ : List[str] = patch_size
UpperCAmelCase_ : Tuple = max_length
UpperCAmelCase_ : Optional[Any] = num_mel_bins
UpperCAmelCase_ : List[str] = is_training
UpperCAmelCase_ : List[Any] = use_labels
UpperCAmelCase_ : Dict = hidden_size
UpperCAmelCase_ : List[str] = num_hidden_layers
UpperCAmelCase_ : List[Any] = num_attention_heads
UpperCAmelCase_ : Union[str, Any] = intermediate_size
UpperCAmelCase_ : Optional[Any] = hidden_act
UpperCAmelCase_ : Optional[int] = hidden_dropout_prob
UpperCAmelCase_ : str = attention_probs_dropout_prob
UpperCAmelCase_ : str = type_sequence_label_size
UpperCAmelCase_ : Optional[int] = initializer_range
UpperCAmelCase_ : Any = scope
UpperCAmelCase_ : str = frequency_stride
UpperCAmelCase_ : Dict = time_stride
# in AST, the seq length equals the number of patches + 2 (we add 2 for the [CLS] and distillation tokens)
UpperCAmelCase_ : Any = (self.num_mel_bins - self.patch_size) // self.frequency_stride + 1
UpperCAmelCase_ : List[str] = (self.max_length - self.patch_size) // self.time_stride + 1
UpperCAmelCase_ : Optional[int] = frequency_out_dimension * time_out_dimension
UpperCAmelCase_ : Union[str, Any] = num_patches + 2
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ : Optional[Any] = floats_tensor([self.batch_size, self.max_length, self.num_mel_bins] )
UpperCAmelCase_ : Optional[Any] = None
if self.use_labels:
UpperCAmelCase_ : Dict = ids_tensor([self.batch_size] , self.type_sequence_label_size )
UpperCAmelCase_ : Optional[Any] = self.get_config()
return config, input_values, labels
def UpperCamelCase__ ( self ):
"""simple docstring"""
return ASTConfig(
patch_size=self.patch_size , max_length=self.max_length , num_mel_bins=self.num_mel_bins , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , is_decoder=lowercase_ , initializer_range=self.initializer_range , frequency_stride=self.frequency_stride , time_stride=self.time_stride , )
def UpperCamelCase__ ( self , lowercase_ , lowercase_ , lowercase_ ):
"""simple docstring"""
UpperCAmelCase_ : Dict = ASTModel(config=lowercase_ )
model.to(lowercase_ )
model.eval()
UpperCAmelCase_ : str = model(lowercase_ )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ : Union[str, Any] = self.prepare_config_and_inputs()
(
(
UpperCAmelCase_
) , (
UpperCAmelCase_
) , (
UpperCAmelCase_
) ,
) : Optional[int] = config_and_inputs
UpperCAmelCase_ : str = {"input_values": input_values}
return config, inputs_dict
@require_torch
class A_ (lowercase__ ,lowercase__ ,unittest.TestCase ):
'''simple docstring'''
SCREAMING_SNAKE_CASE__ : int = (
(
ASTModel,
ASTForAudioClassification,
)
if is_torch_available()
else ()
)
SCREAMING_SNAKE_CASE__ : Dict = (
{"""audio-classification""": ASTForAudioClassification, """feature-extraction""": ASTModel}
if is_torch_available()
else {}
)
SCREAMING_SNAKE_CASE__ : Optional[int] = False
SCREAMING_SNAKE_CASE__ : Tuple = False
SCREAMING_SNAKE_CASE__ : Optional[Any] = False
SCREAMING_SNAKE_CASE__ : Tuple = False
def UpperCamelCase__ ( self , lowercase_ , lowercase_ , lowercase_ , lowercase_ , lowercase_ ):
"""simple docstring"""
if pipeline_test_casse_name == "AudioClassificationPipelineTests":
return True
return False
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ : Optional[int] = ASTModelTester(self )
UpperCAmelCase_ : Tuple = ConfigTester(self , config_class=lowercase_ , has_text_modality=lowercase_ , hidden_size=37 )
def UpperCamelCase__ ( self ):
"""simple docstring"""
self.config_tester.run_common_tests()
@unittest.skip(reason="AST does not use inputs_embeds" )
def UpperCamelCase__ ( self ):
"""simple docstring"""
pass
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ : Union[str, Any] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ : Optional[Any] = model_class(lowercase_ )
self.assertIsInstance(model.get_input_embeddings() , (nn.Module) )
UpperCAmelCase_ : Optional[Any] = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(lowercase_ , nn.Linear ) )
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ , UpperCAmelCase_ : List[Any] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
UpperCAmelCase_ : Dict = model_class(lowercase_ )
UpperCAmelCase_ : str = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
UpperCAmelCase_ : int = [*signature.parameters.keys()]
UpperCAmelCase_ : List[Any] = ["input_values"]
self.assertListEqual(arg_names[:1] , lowercase_ )
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ : Union[str, Any] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*lowercase_ )
@slow
def UpperCamelCase__ ( self ):
"""simple docstring"""
for model_name in AUDIO_SPECTROGRAM_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
UpperCAmelCase_ : Optional[Any] = ASTModel.from_pretrained(lowercase_ )
self.assertIsNotNone(lowercase_ )
def __a ( ):
UpperCAmelCase_ : Any = hf_hub_download(
repo_id="nielsr/audio-spectogram-transformer-checkpoint", filename="sample_audio.flac", repo_type="dataset" )
UpperCAmelCase_ , UpperCAmelCase_ : Any = torchaudio.load(__lowerCamelCase )
return audio, sampling_rate
@require_torch
@require_torchaudio
class A_ (unittest.TestCase ):
'''simple docstring'''
@cached_property
def UpperCamelCase__ ( self ):
"""simple docstring"""
return (
ASTFeatureExtractor.from_pretrained("MIT/ast-finetuned-audioset-10-10-0.4593" )
if is_torchaudio_available()
else None
)
@slow
def UpperCamelCase__ ( self ):
"""simple docstring"""
UpperCAmelCase_ : str = self.default_feature_extractor
UpperCAmelCase_ : str = ASTForAudioClassification.from_pretrained("MIT/ast-finetuned-audioset-10-10-0.4593" ).to(lowercase_ )
UpperCAmelCase_ : List[Any] = self.default_feature_extractor
UpperCAmelCase_ , UpperCAmelCase_ : Union[str, Any] = prepare_audio()
UpperCAmelCase_ : int = audio.squeeze().numpy()
UpperCAmelCase_ : Any = feature_extractor(lowercase_ , sampling_rate=lowercase_ , return_tensors="pt" ).to(lowercase_ )
# forward pass
with torch.no_grad():
UpperCAmelCase_ : Optional[Any] = model(**lowercase_ )
# verify the logits
UpperCAmelCase_ : List[Any] = torch.Size((1, 527) )
self.assertEqual(outputs.logits.shape , lowercase_ )
UpperCAmelCase_ : List[Any] = torch.tensor([-0.87_60, -7.00_42, -8.66_02] ).to(lowercase_ )
self.assertTrue(torch.allclose(outputs.logits[0, :3] , lowercase_ , atol=1E-4 ) )
| 61 |
"""simple docstring"""
import argparse
import json
import os
import sys
import tempfile
import unittest
from argparse import Namespace
from dataclasses import dataclass, field
from enum import Enum
from pathlib import Path
from typing import List, Literal, Optional
import yaml
from transformers import HfArgumentParser, TrainingArguments
from transformers.hf_argparser import make_choice_type_function, string_to_bool
# Since Python 3.10, we can use the builtin `|` operator for Union types
# See PEP 604: https://peps.python.org/pep-0604
__SCREAMING_SNAKE_CASE : int = sys.version_info >= (3, 10)
def _a ( _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
return field(default_factory=lambda: default , metadata=_SCREAMING_SNAKE_CASE )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: float
__lowercase: str
__lowercase: bool
@dataclass
class __A :
'''simple docstring'''
__lowercase: int = 42
__lowercase: str = field(default="""toto""" , metadata={"""help""": """help message"""})
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: Optional[bool] = None
class __A (snake_case__):
'''simple docstring'''
__lowercase: str = """titi"""
__lowercase: Any = """toto"""
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """titi"""
__lowercase: Optional[Any] = """toto"""
__lowercase: List[Any] = 42
@dataclass
class __A :
'''simple docstring'''
__lowercase: BasicEnum = "toto"
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
snake_case_ = BasicEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: MixedTypeEnum = "toto"
def lowerCAmelCase ( self : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = MixedTypeEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: Optional[int] = None
__lowercase: Optional[float] = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: Optional[str] = None
__lowercase: Optional[List[str]] = list_field(default=[])
__lowercase: Optional[List[int]] = list_field(default=[])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = list_field(default=[])
__lowercase: List[int] = list_field(default=[1, 2, 3])
__lowercase: List[str] = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
__lowercase: List[float] = list_field(default=[0.1, 0.2, 0.3])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = field()
__lowercase: str = field()
__lowercase: BasicEnum = field()
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
snake_case_ = BasicEnum(self.required_enum )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: "BasicEnum" = field()
__lowercase: "Optional[bool]" = None
__lowercase: "str" = field(default="""toto""" , metadata={"""help""": """help message"""})
__lowercase: "List[str]" = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
if is_python_no_less_than_3_10:
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: bool | None = None
@dataclass
class __A :
'''simple docstring'''
__lowercase: int | None = None
__lowercase: float | None = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: str | None = None
__lowercase: list[str] | None = list_field(default=[])
__lowercase: list[int] | None = list_field(default=[])
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : argparse.ArgumentParser , UpperCAmelCase_ : argparse.ArgumentParser ) ->Optional[int]:
"""simple docstring"""
self.assertEqual(len(a._actions ) , len(b._actions ) )
for x, y in zip(a._actions , b._actions ):
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
# Choices with mixed type have custom function as "type"
# So we need to compare results directly for equality
if xx.get("""choices""" , UpperCAmelCase_ ) and yy.get("""choices""" , UpperCAmelCase_ ):
for expected_choice in yy["choices"] + xx["choices"]:
self.assertEqual(xx["""type"""](UpperCAmelCase_ ) , yy["""type"""](UpperCAmelCase_ ) )
del xx["type"], yy["type"]
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--bar""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--flag""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = ["""--foo""", """1""", """--baz""", """quux""", """--bar""", """0.5"""]
((snake_case_) , ) = parser.parse_args_into_dataclasses(UpperCAmelCase_ , look_for_args_file=UpperCAmelCase_ )
self.assertFalse(example.flag )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=42 , type=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
# A boolean no_* argument always has to come after its "default: True" regular counter-part
# and its default must be set to False
expected.add_argument("""--no_baz""" , action="""store_false""" , default=UpperCAmelCase_ , dest="""baz""" )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
snake_case_ = [WithDefaultBoolExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--no_baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """True""", """--baz""", """True""", """--opt""", """True"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """False""", """--baz""", """False""", """--opt""", """False"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
def lowerCAmelCase ( self : int ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=["""titi""", """toto""", 42] , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args_into_dataclasses([] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.toto )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """titi"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.titi )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """42"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.fourtytwo )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
@dataclass
class __A :
'''simple docstring'''
__lowercase: Literal["titi", "toto", 42] = "toto"
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=("""titi""", """toto""", 42) , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo_int""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--bar_int""" , nargs="""+""" , default=[1, 2, 3] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_float""" , nargs="""+""" , default=[0.1, 0.2, 0.3] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(
UpperCAmelCase_ , Namespace(foo_int=[] , bar_int=[1, 2, 3] , foo_str=["""Hallo""", """Bonjour""", """Hello"""] , foo_float=[0.1, 0.2, 0.3] ) , )
snake_case_ = parser.parse_args("""--foo_int 1 --bar_int 2 3 --foo_str a b c --foo_float 0.1 0.7""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo_int=[1] , bar_int=[2, 3] , foo_str=["""a""", """b""", """c"""] , foo_float=[0.1, 0.7] ) )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--bar""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--baz""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--ces""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--des""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
snake_case_ = [OptionalExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , bar=UpperCAmelCase_ , baz=UpperCAmelCase_ , ces=[] , des=[] ) )
snake_case_ = parser.parse_args("""--foo 12 --bar 3.14 --baz 42 --ces a b c --des 1 2 3""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=12 , bar=3.14 , baz="""42""" , ces=["""a""", """b""", """c"""] , des=[1, 2, 3] ) )
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--required_list""" , nargs="""+""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--required_str""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
snake_case_ = parser.parse_dict(UpperCAmelCase_ )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
"""extra""": 42,
}
self.assertRaises(UpperCAmelCase_ , parser.parse_dict , UpperCAmelCase_ , allow_extra_keys=UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_json""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.json""" , """w+""" ) as f:
json.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.json""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_yaml""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.yaml""" , """w+""" ) as f:
yaml.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.yaml""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
_A = '\n# Installazione di Transformers\n! pip install transformers datasets\n# Per installare dalla fonte invece dell\'ultima versione rilasciata, commenta il comando sopra e\n# rimuovi la modalità commento al comando seguente.\n# ! pip install git+https://github.com/huggingface/transformers.git\n'
_A = [{'type': 'code', 'content': INSTALL_CONTENT}]
_A = {
'{processor_class}': 'FakeProcessorClass',
'{model_class}': 'FakeModelClass',
'{object_class}': 'FakeObjectClass',
}
| 62 |
"""simple docstring"""
import logging
import os
from typing import Dict, List, Optional, Union
import torch
import torch.nn as nn
from accelerate.utils.imports import (
is_abit_bnb_available,
is_abit_bnb_available,
is_bnb_available,
)
from ..big_modeling import dispatch_model, init_empty_weights
from .dataclasses import BnbQuantizationConfig
from .modeling import (
find_tied_parameters,
get_balanced_memory,
infer_auto_device_map,
load_checkpoint_in_model,
offload_weight,
set_module_tensor_to_device,
)
if is_bnb_available():
import bitsandbytes as bnb
from copy import deepcopy
__SCREAMING_SNAKE_CASE : Any = logging.getLogger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = False , ) -> Optional[Any]:
snake_case_ = bnb_quantization_config.load_in_abit
snake_case_ = bnb_quantization_config.load_in_abit
if load_in_abit and not is_abit_bnb_available():
raise ImportError(
"""You have a version of `bitsandbytes` that is not compatible with 8bit quantization,"""
""" make sure you have the latest version of `bitsandbytes` installed.""" )
if load_in_abit and not is_abit_bnb_available():
raise ValueError(
"""You have a version of `bitsandbytes` that is not compatible with 4bit quantization,"""
"""make sure you have the latest version of `bitsandbytes` installed.""" )
snake_case_ = []
# custom device map
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) and len(device_map.keys() ) > 1:
snake_case_ = [key for key, value in device_map.items() if value in ["""disk""", """cpu"""]]
# We keep some modules such as the lm_head in their original dtype for numerical stability reasons
if bnb_quantization_config.skip_modules is None:
snake_case_ = get_keys_to_not_convert(_SCREAMING_SNAKE_CASE )
# add cpu modules to skip modules only for 4-bit modules
if load_in_abit:
bnb_quantization_config.skip_modules.extend(_SCREAMING_SNAKE_CASE )
snake_case_ = bnb_quantization_config.skip_modules
# We add the modules we want to keep in full precision
if bnb_quantization_config.keep_in_fpaa_modules is None:
snake_case_ = []
snake_case_ = bnb_quantization_config.keep_in_fpaa_modules
modules_to_not_convert.extend(_SCREAMING_SNAKE_CASE )
# compatibility with peft
snake_case_ = load_in_abit
snake_case_ = load_in_abit
snake_case_ = get_parameter_device(_SCREAMING_SNAKE_CASE )
if model_device.type != "meta":
# quantization of an already loaded model
logger.warning(
"""It is not recommended to quantize a loaded model. """
"""The model should be instantiated under the `init_empty_weights` context manager.""" )
snake_case_ = replace_with_bnb_layers(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
# convert param to the right dtype
snake_case_ = bnb_quantization_config.torch_dtype
for name, param in model.state_dict().items():
if any(module_to_keep_in_fpaa in name for module_to_keep_in_fpaa in keep_in_fpaa_modules ):
param.to(torch.floataa )
if param.dtype != torch.floataa:
snake_case_ = name.replace(""".weight""" , """""" ).replace(""".bias""" , """""" )
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if param is not None:
param.to(torch.floataa )
elif torch.is_floating_point(_SCREAMING_SNAKE_CASE ):
param.to(_SCREAMING_SNAKE_CASE )
if model_device.type == "cuda":
# move everything to cpu in the first place because we can't do quantization if the weights are already on cuda
model.cuda(torch.cuda.current_device() )
torch.cuda.empty_cache()
elif torch.cuda.is_available():
model.to(torch.cuda.current_device() )
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info(
f"""The model device type is {model_device.type}. However, cuda is needed for quantization."""
"""We move the model to cuda.""" )
return model
elif weights_location is None:
raise RuntimeError(
f"""`weights_location` needs to be the folder path containing the weights of the model, but we found {weights_location} """ )
else:
with init_empty_weights():
snake_case_ = replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
snake_case_ = get_quantized_model_device_map(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , max_memory=_SCREAMING_SNAKE_CASE , no_split_module_classes=_SCREAMING_SNAKE_CASE , )
if offload_state_dict is None and device_map is not None and "disk" in device_map.values():
snake_case_ = True
snake_case_ = any(x in list(device_map.values() ) for x in ["""cpu""", """disk"""] )
load_checkpoint_in_model(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , dtype=bnb_quantization_config.torch_dtype , offload_folder=_SCREAMING_SNAKE_CASE , offload_state_dict=_SCREAMING_SNAKE_CASE , keep_in_fpaa_modules=bnb_quantization_config.keep_in_fpaa_modules , offload_abit_bnb=load_in_abit and offload , )
return dispatch_model(_SCREAMING_SNAKE_CASE , device_map=_SCREAMING_SNAKE_CASE , offload_dir=_SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if device_map is None:
if torch.cuda.is_available():
snake_case_ = {"""""": torch.cuda.current_device()}
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info("""The device_map was not initialized.""" """Setting device_map to `{'':torch.cuda.current_device()}`.""" )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
if device_map not in ["auto", "balanced", "balanced_low_0", "sequential"]:
raise ValueError(
"""If passing a string for `device_map`, please choose 'auto', 'balanced', 'balanced_low_0' or """
"""'sequential'.""" )
snake_case_ = {}
special_dtypes.update(
{
name: bnb_quantization_config.torch_dtype
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.skip_modules )
} )
special_dtypes.update(
{
name: torch.floataa
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.keep_in_fpaa_modules )
} )
snake_case_ = {}
snake_case_ = special_dtypes
snake_case_ = no_split_module_classes
snake_case_ = bnb_quantization_config.target_dtype
# get max_memory for each device.
if device_map != "sequential":
snake_case_ = get_balanced_memory(
_SCREAMING_SNAKE_CASE , low_zero=(device_map == """balanced_low_0""") , max_memory=_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE , )
snake_case_ = max_memory
snake_case_ = infer_auto_device_map(_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
# check if don't have any quantized module on the cpu
snake_case_ = bnb_quantization_config.skip_modules + bnb_quantization_config.keep_in_fpaa_modules
snake_case_ = {
key: device_map[key] for key in device_map.keys() if key not in modules_not_to_convert
}
for device in ["cpu", "disk"]:
if device in device_map_without_some_modules.values():
if bnb_quantization_config.load_in_abit:
raise ValueError(
"""
Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit
the quantized model. If you want to dispatch the model on the CPU or the disk while keeping
these modules in `torch_dtype`, you need to pass a custom `device_map` to
`load_and_quantize_model`. Check
https://huggingface.co/docs/accelerate/main/en/usage_guides/quantization#offload-modules-to-cpu-and-disk
for more details.
""" )
else:
logger.info(
"""Some modules are are offloaded to the CPU or the disk. Note that these modules will be converted to 8-bit""" )
del device_map_without_some_modules
return device_map
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if modules_to_not_convert is None:
snake_case_ = []
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not has_been_replaced:
logger.warning(
"""You are loading your model in 8bit or 4bit but no linear modules were found in your model."""
""" this can happen for some architectures such as gpt2 that uses Conv1D instead of Linear layers."""
""" Please double check your model architecture, or submit an issue on github if you think this is"""
""" a bug.""" )
return model
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , ) -> List[Any]:
snake_case_ = False
for name, module in model.named_children():
if current_key_name is None:
snake_case_ = []
current_key_name.append(_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , nn.Linear ) and name not in modules_to_not_convert:
# Check if the current key is not in the `modules_to_not_convert`
snake_case_ = """.""".join(_SCREAMING_SNAKE_CASE )
snake_case_ = True
for key in modules_to_not_convert:
if (
(key in current_key_name_str) and (key + "." in current_key_name_str)
) or key == current_key_name_str:
snake_case_ = False
break
if proceed:
# Load bnb module with empty weight and replace ``nn.Linear` module
if bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.LinearabitLt(
module.in_features , module.out_features , module.bias is not None , has_fpaa_weights=_SCREAMING_SNAKE_CASE , threshold=bnb_quantization_config.llm_inta_threshold , )
elif bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.Linearabit(
module.in_features , module.out_features , module.bias is not None , bnb_quantization_config.bnb_abit_compute_dtype , compress_statistics=bnb_quantization_config.bnb_abit_use_double_quant , quant_type=bnb_quantization_config.bnb_abit_quant_type , )
else:
raise ValueError("""load_in_8bit and load_in_4bit can't be both False""" )
snake_case_ = module.weight.data
if module.bias is not None:
snake_case_ = module.bias.data
bnb_module.requires_grad_(_SCREAMING_SNAKE_CASE )
setattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = True
if len(list(module.children() ) ) > 0:
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = has_been_replaced | _has_been_replaced
# Remove the last key for recursion
current_key_name.pop(-1 )
return model, has_been_replaced
def _a ( _SCREAMING_SNAKE_CASE ) -> Any:
# Create a copy of the model
with init_empty_weights():
snake_case_ = deepcopy(_SCREAMING_SNAKE_CASE ) # this has 0 cost since it is done inside `init_empty_weights` context manager`
snake_case_ = find_tied_parameters(_SCREAMING_SNAKE_CASE )
# For compatibility with Accelerate < 0.18
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
snake_case_ = sum(list(tied_params.values() ) , [] ) + list(tied_params.keys() )
else:
snake_case_ = sum(_SCREAMING_SNAKE_CASE , [] )
snake_case_ = len(_SCREAMING_SNAKE_CASE ) > 0
# Check if it is a base model
snake_case_ = False
if hasattr(_SCREAMING_SNAKE_CASE , """base_model_prefix""" ):
snake_case_ = not hasattr(_SCREAMING_SNAKE_CASE , model.base_model_prefix )
# Ignore this for base models (BertModel, GPT2Model, etc.)
if (not has_tied_params) and is_base_model:
return []
# otherwise they have an attached head
snake_case_ = list(model.named_children() )
snake_case_ = [list_modules[-1][0]]
# add last module together with tied weights
snake_case_ = set(_SCREAMING_SNAKE_CASE ) - set(_SCREAMING_SNAKE_CASE )
snake_case_ = list(set(_SCREAMING_SNAKE_CASE ) ) + list(_SCREAMING_SNAKE_CASE )
# remove ".weight" from the keys
snake_case_ = [""".weight""", """.bias"""]
snake_case_ = []
for name in list_untouched:
for name_to_remove in names_to_remove:
if name_to_remove in name:
snake_case_ = name.replace(_SCREAMING_SNAKE_CASE , """""" )
filtered_module_names.append(_SCREAMING_SNAKE_CASE )
return filtered_module_names
def _a ( _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for m in model.modules():
if isinstance(_SCREAMING_SNAKE_CASE , bnb.nn.Linearabit ):
return True
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]:
return next(parameter.parameters() ).device
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
# if it is not quantized, we quantize and offload the quantized weights and the SCB stats
if fpaa_statistics is None:
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , 0 , dtype=_SCREAMING_SNAKE_CASE , value=_SCREAMING_SNAKE_CASE )
snake_case_ = param_name
snake_case_ = model
if "." in tensor_name:
snake_case_ = tensor_name.split(""".""" )
for split in splits[:-1]:
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if new_module is None:
raise ValueError(f"""{module} has no attribute {split}.""" )
snake_case_ = new_module
snake_case_ = splits[-1]
# offload weights
snake_case_ = False
offload_weight(module._parameters[tensor_name] , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
if hasattr(module._parameters[tensor_name] , """SCB""" ):
offload_weight(
module._parameters[tensor_name].SCB , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE , )
else:
offload_weight(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
offload_weight(_SCREAMING_SNAKE_CASE , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , """meta""" , dtype=_SCREAMING_SNAKE_CASE , value=torch.empty(*param.size() ) )
| 347 | 0 |
'''simple docstring'''
import unittest
from transformers import DebertaConfig, is_torch_available
from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import (
DebertaForMaskedLM,
DebertaForQuestionAnswering,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaModel,
)
from transformers.models.deberta.modeling_deberta import DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST
class __SCREAMING_SNAKE_CASE (lowerCamelCase_ ):
"""simple docstring"""
def __init__( self : List[Any] , __a : int , __a : Optional[Any]=13 , __a : str=7 , __a : Optional[int]=True , __a : List[Any]=True , __a : Any=True , __a : List[str]=True , __a : str=99 , __a : int=32 , __a : Any=5 , __a : Union[str, Any]=4 , __a : Optional[int]=37 , __a : Optional[Any]="gelu" , __a : Any=0.1 , __a : str=0.1 , __a : Any=5_12 , __a : Optional[Any]=16 , __a : Dict=2 , __a : Union[str, Any]=0.02 , __a : Any=False , __a : Optional[int]=True , __a : List[Any]="None" , __a : Optional[int]=3 , __a : Dict=4 , __a : List[str]=None , ):
_a = parent
_a = batch_size
_a = seq_length
_a = is_training
_a = use_input_mask
_a = use_token_type_ids
_a = use_labels
_a = vocab_size
_a = hidden_size
_a = num_hidden_layers
_a = num_attention_heads
_a = intermediate_size
_a = hidden_act
_a = hidden_dropout_prob
_a = attention_probs_dropout_prob
_a = max_position_embeddings
_a = type_vocab_size
_a = type_sequence_label_size
_a = initializer_range
_a = num_labels
_a = num_choices
_a = relative_attention
_a = position_biased_input
_a = pos_att_type
_a = scope
def UpperCamelCase__ ( self : Optional[int] ):
_a = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
_a = None
if self.use_input_mask:
_a = ids_tensor([self.batch_size, self.seq_length] , vocab_size=2 )
_a = None
if self.use_token_type_ids:
_a = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
_a = None
_a = None
_a = None
if self.use_labels:
_a = ids_tensor([self.batch_size] , self.type_sequence_label_size )
_a = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
_a = ids_tensor([self.batch_size] , self.num_choices )
_a = self.get_config()
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def UpperCamelCase__ ( self : Optional[int] ):
return DebertaConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , initializer_range=self.initializer_range , relative_attention=self.relative_attention , position_biased_input=self.position_biased_input , pos_att_type=self.pos_att_type , )
def UpperCamelCase__ ( self : Any ):
_a = self.get_config()
_a = 3_00
return config
def UpperCamelCase__ ( self : List[str] , __a : Dict ):
self.parent.assertListEqual(list(result.loss.size() ) , [] )
def UpperCamelCase__ ( self : List[Any] , __a : int , __a : Dict , __a : Tuple , __a : str , __a : Union[str, Any] , __a : List[str] , __a : List[Any] ):
_a = DebertaModel(config=__a )
model.to(__a )
model.eval()
_a = model(__a , attention_mask=__a , token_type_ids=__a )[0]
_a = model(__a , token_type_ids=__a )[0]
_a = model(__a )[0]
self.parent.assertListEqual(list(sequence_output.size() ) , [self.batch_size, self.seq_length, self.hidden_size] )
def UpperCamelCase__ ( self : str , __a : List[str] , __a : Optional[Any] , __a : Tuple , __a : List[Any] , __a : Dict , __a : Optional[Any] , __a : List[str] ):
_a = DebertaForMaskedLM(config=__a )
model.to(__a )
model.eval()
_a = model(__a , attention_mask=__a , token_type_ids=__a , labels=__a )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
def UpperCamelCase__ ( self : Optional[Any] , __a : int , __a : Optional[int] , __a : Dict , __a : int , __a : Optional[int] , __a : str , __a : Dict ):
_a = self.num_labels
_a = DebertaForSequenceClassification(__a )
model.to(__a )
model.eval()
_a = model(__a , attention_mask=__a , token_type_ids=__a , labels=__a )
self.parent.assertListEqual(list(result.logits.size() ) , [self.batch_size, self.num_labels] )
self.check_loss_output(__a )
def UpperCamelCase__ ( self : List[str] , __a : int , __a : Dict , __a : Union[str, Any] , __a : Dict , __a : str , __a : Optional[Any] , __a : List[str] ):
_a = self.num_labels
_a = DebertaForTokenClassification(config=__a )
model.to(__a )
model.eval()
_a = model(__a , attention_mask=__a , token_type_ids=__a , labels=__a )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) )
def UpperCamelCase__ ( self : Optional[Any] , __a : List[Any] , __a : Dict , __a : List[str] , __a : Optional[int] , __a : Union[str, Any] , __a : int , __a : Optional[Any] ):
_a = DebertaForQuestionAnswering(config=__a )
model.to(__a )
model.eval()
_a = model(
__a , attention_mask=__a , token_type_ids=__a , start_positions=__a , end_positions=__a , )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
def UpperCamelCase__ ( self : List[str] ):
_a = self.prepare_config_and_inputs()
(
(
_a
) , (
_a
) , (
_a
) , (
_a
) , (
_a
) , (
_a
) , (
_a
) ,
) = config_and_inputs
_a = {"input_ids": input_ids, "token_type_ids": token_type_ids, "attention_mask": input_mask}
return config, inputs_dict
@require_torch
class __SCREAMING_SNAKE_CASE (lowerCamelCase_ , lowerCamelCase_ , unittest.TestCase ):
"""simple docstring"""
__a =(
(
DebertaModel,
DebertaForMaskedLM,
DebertaForSequenceClassification,
DebertaForTokenClassification,
DebertaForQuestionAnswering,
)
if is_torch_available()
else ()
)
__a =(
{
'feature-extraction': DebertaModel,
'fill-mask': DebertaForMaskedLM,
'question-answering': DebertaForQuestionAnswering,
'text-classification': DebertaForSequenceClassification,
'token-classification': DebertaForTokenClassification,
'zero-shot': DebertaForSequenceClassification,
}
if is_torch_available()
else {}
)
__a =True
__a =False
__a =False
__a =False
__a =False
def UpperCamelCase__ ( self : Tuple ):
_a = DebertaModelTester(self )
_a = ConfigTester(self , config_class=__a , hidden_size=37 )
def UpperCamelCase__ ( self : Optional[int] ):
self.config_tester.run_common_tests()
def UpperCamelCase__ ( self : str ):
_a = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_deberta_model(*__a )
def UpperCamelCase__ ( self : Any ):
_a = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_deberta_for_sequence_classification(*__a )
def UpperCamelCase__ ( self : List[str] ):
_a = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_deberta_for_masked_lm(*__a )
def UpperCamelCase__ ( self : List[Any] ):
_a = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_deberta_for_question_answering(*__a )
def UpperCamelCase__ ( self : Any ):
_a = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_deberta_for_token_classification(*__a )
@slow
def UpperCamelCase__ ( self : List[str] ):
for model_name in DEBERTA_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
_a = DebertaModel.from_pretrained(__a )
self.assertIsNotNone(__a )
@require_torch
@require_sentencepiece
@require_tokenizers
class __SCREAMING_SNAKE_CASE (unittest.TestCase ):
"""simple docstring"""
@unittest.skip(reason="Model not available yet" )
def UpperCamelCase__ ( self : Tuple ):
pass
@slow
def UpperCamelCase__ ( self : List[Any] ):
_a = DebertaModel.from_pretrained("microsoft/deberta-base" )
_a = torch.tensor([[0, 3_14_14, 2_32, 3_28, 7_40, 11_40, 1_26_95, 69, 4_60_78, 15_88, 2]] )
_a = torch.tensor([[0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]] )
with torch.no_grad():
_a = model(__a , attention_mask=__a )[0]
# compare the actual values for a slice.
_a = torch.tensor(
[[[-0.5986, -0.8055, -0.8462], [1.4484, -0.9348, -0.8059], [0.3123, 0.0032, -1.4131]]] )
self.assertTrue(torch.allclose(output[:, 1:4, 1:4] , __a , atol=1e-4 ) , f'{output[:, 1:4, 1:4]}' )
| 63 |
"""simple docstring"""
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__SCREAMING_SNAKE_CASE : str = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'microsoft/beit-base-patch16-224-pt22k': (
'https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json'
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = """beit"""
def __init__( self : List[str] , UpperCAmelCase_ : List[Any]=8_192 , UpperCAmelCase_ : Dict=768 , UpperCAmelCase_ : int=12 , UpperCAmelCase_ : Tuple=12 , UpperCAmelCase_ : List[Any]=3_072 , UpperCAmelCase_ : Tuple="gelu" , UpperCAmelCase_ : Dict=0.0 , UpperCAmelCase_ : List[str]=0.0 , UpperCAmelCase_ : Any=0.02 , UpperCAmelCase_ : Optional[Any]=1E-12 , UpperCAmelCase_ : int=224 , UpperCAmelCase_ : Tuple=16 , UpperCAmelCase_ : List[str]=3 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[str]=False , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[Any]=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Any=True , UpperCAmelCase_ : Dict=[3, 5, 7, 11] , UpperCAmelCase_ : Tuple=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : List[Any]=0.4 , UpperCAmelCase_ : Optional[Any]=256 , UpperCAmelCase_ : Optional[Any]=1 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : List[str] , ) ->Optional[Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = initializer_range
snake_case_ = layer_norm_eps
snake_case_ = image_size
snake_case_ = patch_size
snake_case_ = num_channels
snake_case_ = use_mask_token
snake_case_ = use_absolute_position_embeddings
snake_case_ = use_relative_position_bias
snake_case_ = use_shared_relative_position_bias
snake_case_ = layer_scale_init_value
snake_case_ = drop_path_rate
snake_case_ = use_mean_pooling
# decode head attributes (semantic segmentation)
snake_case_ = out_indices
snake_case_ = pool_scales
# auxiliary head attributes (semantic segmentation)
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = semantic_loss_ignore_index
class __A (snake_case__):
'''simple docstring'''
__lowercase: List[Any] = version.parse("""1.11""")
@property
def lowerCAmelCase ( self : Dict ) ->Mapping[str, Mapping[int, str]]:
"""simple docstring"""
return OrderedDict(
[
("""pixel_values""", {0: """batch""", 1: """num_channels""", 2: """height""", 3: """width"""}),
] )
@property
def lowerCAmelCase ( self : Any ) ->float:
"""simple docstring"""
return 1E-4
| 347 | 0 |
"""simple docstring"""
def UpperCAmelCase__ (snake_case__ : list[int] , snake_case__ : list[int] ):
"""simple docstring"""
if not len(snake_case__ ) == len(snake_case__ ) == 3:
raise ValueError("""Please enter a valid equation.""" )
if equationa[0] == equationa[1] == equationa[0] == equationa[1] == 0:
raise ValueError("""Both a & b of two equations can't be zero.""" )
# Extract the coefficients
_snake_case , _snake_case , _snake_case : Optional[Any] = equationa
_snake_case , _snake_case , _snake_case : Optional[Any] = equationa
# Calculate the determinants of the matrices
_snake_case : Any = aa * ba - aa * ba
_snake_case : Optional[int] = ca * ba - ca * ba
_snake_case : Tuple = aa * ca - aa * ca
# Check if the system of linear equations has a solution (using Cramer's rule)
if determinant == 0:
if determinant_x == determinant_y == 0:
raise ValueError("""Infinite solutions. (Consistent system)""" )
else:
raise ValueError("""No solution. (Inconsistent system)""" )
else:
if determinant_x == determinant_y == 0:
# Trivial solution (Inconsistent system)
return (0.0, 0.0)
else:
_snake_case : int = determinant_x / determinant
_snake_case : List[str] = determinant_y / determinant
# Non-Trivial Solution (Consistent system)
return (x, y)
| 64 |
"""simple docstring"""
import os
import re
import warnings
from shutil import copyfile
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
if TYPE_CHECKING:
from ...tokenization_utils_base import TextInput
from ...utils import logging
__SCREAMING_SNAKE_CASE : Union[str, Any] = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : List[Any] = {'vocab_file': 'spiece.model'}
__SCREAMING_SNAKE_CASE : int = {
'vocab_file': {
't5-small': 'https://huggingface.co/t5-small/resolve/main/spiece.model',
't5-base': 'https://huggingface.co/t5-base/resolve/main/spiece.model',
't5-large': 'https://huggingface.co/t5-large/resolve/main/spiece.model',
't5-3b': 'https://huggingface.co/t5-3b/resolve/main/spiece.model',
't5-11b': 'https://huggingface.co/t5-11b/resolve/main/spiece.model',
}
}
# TODO(PVP) - this should be removed in Transformers v5
__SCREAMING_SNAKE_CASE : Dict = {
't5-small': 512,
't5-base': 512,
't5-large': 512,
't5-3b': 512,
't5-11b': 512,
}
__SCREAMING_SNAKE_CASE : Optional[int] = '▁'
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = VOCAB_FILES_NAMES
__lowercase: Any = PRETRAINED_VOCAB_FILES_MAP
__lowercase: Any = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__lowercase: List[str] = ["""input_ids""", """attention_mask"""]
def __init__( self : Optional[int] , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Optional[Any]="</s>" , UpperCAmelCase_ : Optional[Any]="<unk>" , UpperCAmelCase_ : Any="<pad>" , UpperCAmelCase_ : Tuple=100 , UpperCAmelCase_ : Optional[Any]=None , UpperCAmelCase_ : Optional[Dict[str, Any]] = None , UpperCAmelCase_ : Optional[int]=True , **UpperCAmelCase_ : Dict , ) ->None:
"""simple docstring"""
if extra_ids > 0 and additional_special_tokens is None:
snake_case_ = [F"""<extra_id_{i}>""" for i in range(UpperCAmelCase_ )]
elif extra_ids > 0 and additional_special_tokens is not None:
# Check that we have the right number of extra_id special tokens
snake_case_ = len(set(filter(lambda UpperCAmelCase_ : bool("""extra_id""" in str(UpperCAmelCase_ ) ) , UpperCAmelCase_ ) ) )
if extra_tokens != extra_ids:
raise ValueError(
F"""Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are"""
""" provided to T5Tokenizer. In this case the additional_special_tokens must include the extra_ids"""
""" tokens""" )
if legacy:
logger.warning_once(
F"""You are using the legacy behaviour of the {self.__class__}. This means that tokens that come after special tokens will not be properly handled. We recommend you to"""
""" read the related pull request available at https://github.com/huggingface/transformers/pull/24565""" )
snake_case_ = legacy
snake_case_ = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
eos_token=UpperCAmelCase_ , unk_token=UpperCAmelCase_ , pad_token=UpperCAmelCase_ , extra_ids=UpperCAmelCase_ , additional_special_tokens=UpperCAmelCase_ , sp_model_kwargs=self.sp_model_kwargs , legacy=UpperCAmelCase_ , **UpperCAmelCase_ , )
snake_case_ = vocab_file
snake_case_ = extra_ids
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(UpperCAmelCase_ )
@staticmethod
def lowerCAmelCase ( UpperCAmelCase_ : str , UpperCAmelCase_ : Dict , UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
if pretrained_model_name_or_path in TaTokenizer.max_model_input_sizes:
snake_case_ = TaTokenizer.max_model_input_sizes[pretrained_model_name_or_path]
if init_max_model_length is not None and init_max_model_length != max_model_length:
return init_max_model_length
elif init_max_model_length is None:
warnings.warn(
"""This tokenizer was incorrectly instantiated with a model max length of"""
F""" {deprecated_max_model_length} which will be corrected in Transformers v5.\nFor now, this"""
""" behavior is kept to avoid breaking backwards compatibility when padding/encoding with"""
""" `truncation is True`.\n- Be aware that you SHOULD NOT rely on"""
F""" {pretrained_model_name_or_path} automatically truncating your input to"""
F""" {deprecated_max_model_length} when padding/encoding.\n- If you want to encode/pad to sequences"""
F""" longer than {deprecated_max_model_length} you can either instantiate this tokenizer with"""
""" `model_max_length` or pass `max_length` when encoding/padding.\n- To avoid this warning, please"""
""" instantiate this tokenizer with `model_max_length` set to your preferred value.""" , UpperCAmelCase_ , )
return max_model_length
@property
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
return self.sp_model.get_piece_size() + self._extra_ids
def lowerCAmelCase ( self : Any ) ->Optional[int]:
"""simple docstring"""
snake_case_ = {self.convert_ids_to_tokens(UpperCAmelCase_ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None , UpperCAmelCase_ : bool = False ) ->List[int]:
"""simple docstring"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCAmelCase_ , token_ids_a=UpperCAmelCase_ , already_has_special_tokens=UpperCAmelCase_ )
# normal case: some special tokens
if token_ids_a is None:
return ([0] * len(UpperCAmelCase_ )) + [1]
return ([0] * len(UpperCAmelCase_ )) + [1] + ([0] * len(UpperCAmelCase_ )) + [1]
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
return list(
set(filter(lambda UpperCAmelCase_ : bool(re.search(R"""<extra_id_\d+>""" , UpperCAmelCase_ ) ) is not None , self.additional_special_tokens ) ) )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
return [self._convert_token_to_id(UpperCAmelCase_ ) for token in self.get_sentinel_tokens()]
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : List[int] ) ->List[int]:
"""simple docstring"""
if len(UpperCAmelCase_ ) > 0 and token_ids[-1] == self.eos_token_id:
warnings.warn(
F"""This sequence already has {self.eos_token}. In future versions this behavior may lead to duplicated"""
""" eos tokens being added.""" )
return token_ids
else:
return token_ids + [self.eos_token_id]
def lowerCAmelCase ( self : str , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = [self.eos_token_id]
if token_ids_a is None:
return len(token_ids_a + eos ) * [0]
return len(token_ids_a + eos + token_ids_a + eos ) * [0]
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
if token_ids_a is None:
return token_ids_a
else:
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
return token_ids_a + token_ids_a
def __getstate__( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = self.__dict__.copy()
snake_case_ = None
return state
def __setstate__( self : Optional[Any] , UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = d
# for backward compatibility
if not hasattr(self , """sp_model_kwargs""" ):
snake_case_ = {}
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def lowerCAmelCase ( self : int , UpperCAmelCase_ : "TextInput" , **UpperCAmelCase_ : Tuple ) ->List[str]:
"""simple docstring"""
if not self.legacy:
snake_case_ = SPIECE_UNDERLINE + text.replace(UpperCAmelCase_ , """ """ )
return super().tokenize(UpperCAmelCase_ , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Any ) ->Tuple:
"""simple docstring"""
if not self.legacy:
snake_case_ = text.startswith(UpperCAmelCase_ )
if is_first:
snake_case_ = text[1:]
snake_case_ = self.sp_model.encode(UpperCAmelCase_ , out_type=UpperCAmelCase_ )
if not self.legacy and not is_first and not text.startswith(""" """ ) and tokens[0].startswith(UpperCAmelCase_ ):
snake_case_ = ([tokens[0][1:]] if len(tokens[0] ) > 1 else []) + tokens[1:]
return tokens
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
if token.startswith("""<extra_id_""" ):
snake_case_ = re.match(R"""<extra_id_(\d+)>""" , UpperCAmelCase_ )
snake_case_ = int(match.group(1 ) )
return self.vocab_size - num - 1
return self.sp_model.piece_to_id(UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : Optional[Any] ) ->List[Any]:
"""simple docstring"""
if index < self.sp_model.get_piece_size():
snake_case_ = self.sp_model.IdToPiece(UpperCAmelCase_ )
else:
snake_case_ = F"""<extra_id_{self.vocab_size - 1 - index}>"""
return token
def lowerCAmelCase ( self : List[Any] , UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = []
snake_case_ = """"""
snake_case_ = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(UpperCAmelCase_ ) + token
snake_case_ = True
snake_case_ = []
else:
current_sub_tokens.append(UpperCAmelCase_ )
snake_case_ = False
out_string += self.sp_model.decode(UpperCAmelCase_ )
return out_string.strip()
def lowerCAmelCase ( self : str , UpperCAmelCase_ : str , UpperCAmelCase_ : Optional[str] = None ) ->Tuple[str]:
"""simple docstring"""
if not os.path.isdir(UpperCAmelCase_ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case_ = os.path.join(
UpperCAmelCase_ , (filename_prefix + """-""" if filename_prefix else """""") + VOCAB_FILES_NAMES["""vocab_file"""] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(UpperCAmelCase_ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , UpperCAmelCase_ )
elif not os.path.isfile(self.vocab_file ):
with open(UpperCAmelCase_ , """wb""" ) as fi:
snake_case_ = self.sp_model.serialized_model_proto()
fi.write(UpperCAmelCase_ )
return (out_vocab_file,)
| 347 | 0 |
import unittest
from transformers import SqueezeBertConfig, is_torch_available
from transformers.testing_utils import require_sentencepiece, require_tokenizers, require_torch, slow, torch_device
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import (
SQUEEZEBERT_PRETRAINED_MODEL_ARCHIVE_LIST,
SqueezeBertForMaskedLM,
SqueezeBertForMultipleChoice,
SqueezeBertForQuestionAnswering,
SqueezeBertForSequenceClassification,
SqueezeBertForTokenClassification,
SqueezeBertModel,
)
class A ( UpperCAmelCase_ ):
def __init__(self : List[str] , __UpperCAmelCase : int , __UpperCAmelCase : Dict=1_3 , __UpperCAmelCase : str=7 , __UpperCAmelCase : List[Any]=True , __UpperCAmelCase : List[Any]=True , __UpperCAmelCase : Optional[int]=False , __UpperCAmelCase : int=True , __UpperCAmelCase : Dict=9_9 , __UpperCAmelCase : List[Any]=3_2 , __UpperCAmelCase : str=5 , __UpperCAmelCase : List[str]=4 , __UpperCAmelCase : int=6_4 , __UpperCAmelCase : int="gelu" , __UpperCAmelCase : Dict=0.1 , __UpperCAmelCase : str=0.1 , __UpperCAmelCase : Optional[Any]=5_1_2 , __UpperCAmelCase : Optional[Any]=1_6 , __UpperCAmelCase : Any=2 , __UpperCAmelCase : List[Any]=0.02 , __UpperCAmelCase : str=3 , __UpperCAmelCase : Union[str, Any]=4 , __UpperCAmelCase : Optional[int]=None , __UpperCAmelCase : List[Any]=2 , __UpperCAmelCase : List[str]=2 , __UpperCAmelCase : Optional[int]=2 , __UpperCAmelCase : Optional[Any]=2 , __UpperCAmelCase : int=4 , __UpperCAmelCase : Dict=1 , ) -> Any:
"""simple docstring"""
UpperCAmelCase__ = parent
UpperCAmelCase__ = batch_size
UpperCAmelCase__ = seq_length
UpperCAmelCase__ = is_training
UpperCAmelCase__ = use_input_mask
UpperCAmelCase__ = use_token_type_ids
UpperCAmelCase__ = use_labels
UpperCAmelCase__ = vocab_size
UpperCAmelCase__ = hidden_size
UpperCAmelCase__ = num_hidden_layers
UpperCAmelCase__ = num_attention_heads
UpperCAmelCase__ = intermediate_size
UpperCAmelCase__ = hidden_act
UpperCAmelCase__ = hidden_dropout_prob
UpperCAmelCase__ = attention_probs_dropout_prob
UpperCAmelCase__ = max_position_embeddings
UpperCAmelCase__ = type_vocab_size
UpperCAmelCase__ = type_sequence_label_size
UpperCAmelCase__ = initializer_range
UpperCAmelCase__ = num_labels
UpperCAmelCase__ = num_choices
UpperCAmelCase__ = scope
UpperCAmelCase__ = q_groups
UpperCAmelCase__ = k_groups
UpperCAmelCase__ = v_groups
UpperCAmelCase__ = post_attention_groups
UpperCAmelCase__ = intermediate_groups
UpperCAmelCase__ = output_groups
def lowercase_ (self : Union[str, Any] ) -> List[str]:
"""simple docstring"""
UpperCAmelCase__ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
UpperCAmelCase__ = None
if self.use_input_mask:
UpperCAmelCase__ = random_attention_mask([self.batch_size, self.seq_length] )
UpperCAmelCase__ = None
UpperCAmelCase__ = None
UpperCAmelCase__ = None
if self.use_labels:
UpperCAmelCase__ = ids_tensor([self.batch_size] , self.type_sequence_label_size )
UpperCAmelCase__ = ids_tensor([self.batch_size, self.seq_length] , self.num_labels )
UpperCAmelCase__ = ids_tensor([self.batch_size] , self.num_choices )
UpperCAmelCase__ = self.get_config()
return config, input_ids, input_mask, sequence_labels, token_labels, choice_labels
def lowercase_ (self : Optional[int] ) -> Union[str, Any]:
"""simple docstring"""
return SqueezeBertConfig(
embedding_size=self.hidden_size , vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , attention_probs_dropout_prob=self.hidden_dropout_prob , attention_dropout=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , initializer_range=self.initializer_range , q_groups=self.q_groups , k_groups=self.k_groups , v_groups=self.v_groups , post_attention_groups=self.post_attention_groups , intermediate_groups=self.intermediate_groups , output_groups=self.output_groups , )
def lowercase_ (self : Dict , __UpperCAmelCase : List[Any] , __UpperCAmelCase : List[Any] , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Any , __UpperCAmelCase : List[str] ) -> int:
"""simple docstring"""
UpperCAmelCase__ = SqueezeBertModel(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = model(__UpperCAmelCase , __UpperCAmelCase )
UpperCAmelCase__ = model(__UpperCAmelCase )
self.parent.assertEqual(result.last_hidden_state.shape , (self.batch_size, self.seq_length, self.hidden_size) )
def lowercase_ (self : Dict , __UpperCAmelCase : str , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : List[Any] , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : Optional[int] , __UpperCAmelCase : Optional[int] ) -> Optional[Any]:
"""simple docstring"""
UpperCAmelCase__ = SqueezeBertForMaskedLM(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = model(__UpperCAmelCase , attention_mask=__UpperCAmelCase , labels=__UpperCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.vocab_size) )
def lowercase_ (self : Tuple , __UpperCAmelCase : int , __UpperCAmelCase : Optional[int] , __UpperCAmelCase : int , __UpperCAmelCase : Dict , __UpperCAmelCase : Optional[int] , __UpperCAmelCase : Union[str, Any] ) -> int:
"""simple docstring"""
UpperCAmelCase__ = SqueezeBertForQuestionAnswering(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = model(
__UpperCAmelCase , attention_mask=__UpperCAmelCase , start_positions=__UpperCAmelCase , end_positions=__UpperCAmelCase )
self.parent.assertEqual(result.start_logits.shape , (self.batch_size, self.seq_length) )
self.parent.assertEqual(result.end_logits.shape , (self.batch_size, self.seq_length) )
def lowercase_ (self : List[Any] , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : List[str] , __UpperCAmelCase : List[str] , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Tuple , __UpperCAmelCase : List[Any] ) -> List[Any]:
"""simple docstring"""
UpperCAmelCase__ = self.num_labels
UpperCAmelCase__ = SqueezeBertForSequenceClassification(__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = model(__UpperCAmelCase , attention_mask=__UpperCAmelCase , labels=__UpperCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_labels) )
def lowercase_ (self : Dict , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : int , __UpperCAmelCase : str , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : List[str] , __UpperCAmelCase : Union[str, Any] ) -> int:
"""simple docstring"""
UpperCAmelCase__ = self.num_labels
UpperCAmelCase__ = SqueezeBertForTokenClassification(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = model(__UpperCAmelCase , attention_mask=__UpperCAmelCase , labels=__UpperCAmelCase )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.seq_length, self.num_labels) )
def lowercase_ (self : List[str] , __UpperCAmelCase : Optional[Any] , __UpperCAmelCase : Union[str, Any] , __UpperCAmelCase : Tuple , __UpperCAmelCase : Tuple , __UpperCAmelCase : str , __UpperCAmelCase : Any ) -> Optional[Any]:
"""simple docstring"""
UpperCAmelCase__ = self.num_choices
UpperCAmelCase__ = SqueezeBertForMultipleChoice(config=__UpperCAmelCase )
model.to(__UpperCAmelCase )
model.eval()
UpperCAmelCase__ = input_ids.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous()
UpperCAmelCase__ = input_mask.unsqueeze(1 ).expand(-1 , self.num_choices , -1 ).contiguous()
UpperCAmelCase__ = model(
__UpperCAmelCase , attention_mask=__UpperCAmelCase , labels=__UpperCAmelCase , )
self.parent.assertEqual(result.logits.shape , (self.batch_size, self.num_choices) )
def lowercase_ (self : Dict ) -> Optional[int]:
"""simple docstring"""
UpperCAmelCase__ = self.prepare_config_and_inputs()
((UpperCAmelCase__) , (UpperCAmelCase__) , (UpperCAmelCase__) , (UpperCAmelCase__) , (UpperCAmelCase__) , (UpperCAmelCase__)) = config_and_inputs
UpperCAmelCase__ = {"input_ids": input_ids, "attention_mask": input_mask}
return config, inputs_dict
@require_torch
class A ( UpperCAmelCase_ , UpperCAmelCase_ , unittest.TestCase ):
__UpperCAmelCase : Optional[int] = (
(
SqueezeBertModel,
SqueezeBertForMaskedLM,
SqueezeBertForMultipleChoice,
SqueezeBertForQuestionAnswering,
SqueezeBertForSequenceClassification,
SqueezeBertForTokenClassification,
)
if is_torch_available()
else None
)
__UpperCAmelCase : Any = (
{
'feature-extraction': SqueezeBertModel,
'fill-mask': SqueezeBertForMaskedLM,
'question-answering': SqueezeBertForQuestionAnswering,
'text-classification': SqueezeBertForSequenceClassification,
'token-classification': SqueezeBertForTokenClassification,
'zero-shot': SqueezeBertForSequenceClassification,
}
if is_torch_available()
else {}
)
__UpperCAmelCase : str = False
__UpperCAmelCase : Tuple = True
__UpperCAmelCase : Any = False
def lowercase_ (self : Optional[Any] ) -> List[str]:
"""simple docstring"""
UpperCAmelCase__ = SqueezeBertModelTester(self )
UpperCAmelCase__ = ConfigTester(self , config_class=__UpperCAmelCase , dim=3_7 )
def lowercase_ (self : Any ) -> Union[str, Any]:
"""simple docstring"""
self.config_tester.run_common_tests()
def lowercase_ (self : List[str] ) -> Optional[int]:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_model(*__UpperCAmelCase )
def lowercase_ (self : Any ) -> Dict:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_for_masked_lm(*__UpperCAmelCase )
def lowercase_ (self : str ) -> Any:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_for_question_answering(*__UpperCAmelCase )
def lowercase_ (self : Dict ) -> Dict:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_for_sequence_classification(*__UpperCAmelCase )
def lowercase_ (self : Any ) -> int:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_for_token_classification(*__UpperCAmelCase )
def lowercase_ (self : Dict ) -> Any:
"""simple docstring"""
UpperCAmelCase__ = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_squeezebert_for_multiple_choice(*__UpperCAmelCase )
@slow
def lowercase_ (self : List[str] ) -> List[Any]:
"""simple docstring"""
for model_name in SQUEEZEBERT_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
UpperCAmelCase__ = SqueezeBertModel.from_pretrained(__UpperCAmelCase )
self.assertIsNotNone(__UpperCAmelCase )
@require_sentencepiece
@require_tokenizers
@require_torch
class A ( unittest.TestCase ):
@slow
def lowercase_ (self : int ) -> int:
"""simple docstring"""
UpperCAmelCase__ = SqueezeBertForSequenceClassification.from_pretrained("squeezebert/squeezebert-mnli" )
UpperCAmelCase__ = torch.tensor([[1, 2_9_4_1_4, 2_3_2, 3_2_8, 7_4_0, 1_1_4_0, 1_2_6_9_5, 6_9, 1_3, 1_5_8_8, 2]] )
UpperCAmelCase__ = model(__UpperCAmelCase )[0]
UpperCAmelCase__ = torch.Size((1, 3) )
self.assertEqual(output.shape , __UpperCAmelCase )
UpperCAmelCase__ = torch.tensor([[0.6401, -0.0349, -0.6041]] )
self.assertTrue(torch.allclose(__UpperCAmelCase , __UpperCAmelCase , atol=1E-4 ) )
| 65 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE = 1_000_000 ) -> int:
snake_case_ = [i - 1 for i in range(limit + 1 )]
for i in range(2 , limit + 1 ):
if phi[i] == i - 1:
for j in range(2 * i , limit + 1 , _SCREAMING_SNAKE_CASE ):
phi[j] -= phi[j] // i
return sum(phi[2 : limit + 1] )
if __name__ == "__main__":
print(solution())
| 347 | 0 |
"""simple docstring"""
from ...configuration_utils import PretrainedConfig
from ...utils import logging
__a = logging.get_logger(__name__)
__a = {
"uclanlp/visualbert-vqa": "https://huggingface.co/uclanlp/visualbert-vqa/resolve/main/config.json",
"uclanlp/visualbert-vqa-pre": "https://huggingface.co/uclanlp/visualbert-vqa-pre/resolve/main/config.json",
"uclanlp/visualbert-vqa-coco-pre": (
"https://huggingface.co/uclanlp/visualbert-vqa-coco-pre/resolve/main/config.json"
),
"uclanlp/visualbert-vcr": "https://huggingface.co/uclanlp/visualbert-vcr/resolve/main/config.json",
"uclanlp/visualbert-vcr-pre": "https://huggingface.co/uclanlp/visualbert-vcr-pre/resolve/main/config.json",
"uclanlp/visualbert-vcr-coco-pre": (
"https://huggingface.co/uclanlp/visualbert-vcr-coco-pre/resolve/main/config.json"
),
"uclanlp/visualbert-nlvr2": "https://huggingface.co/uclanlp/visualbert-nlvr2/resolve/main/config.json",
"uclanlp/visualbert-nlvr2-pre": "https://huggingface.co/uclanlp/visualbert-nlvr2-pre/resolve/main/config.json",
"uclanlp/visualbert-nlvr2-coco-pre": (
"https://huggingface.co/uclanlp/visualbert-nlvr2-coco-pre/resolve/main/config.json"
)
# See all VisualBERT models at https://huggingface.co/models?filter=visual_bert
}
class lowerCamelCase ( _lowerCAmelCase ):
'''simple docstring'''
_A : int = """visual_bert"""
def __init__( self: List[Any] , snake_case: Union[str, Any]=30_522 , snake_case: Dict=768 , snake_case: Any=512 , snake_case: Any=12 , snake_case: Any=12 , snake_case: List[Any]=3_072 , snake_case: int="gelu" , snake_case: int=0.1 , snake_case: str=0.1 , snake_case: str=512 , snake_case: Dict=2 , snake_case: int=0.0_2 , snake_case: Optional[int]=1E-12 , snake_case: str=False , snake_case: List[Any]=True , snake_case: Union[str, Any]=1 , snake_case: Optional[Any]=0 , snake_case: Tuple=2 , **snake_case: Union[str, Any] , ) -> Union[str, Any]:
super().__init__(pad_token_id=snake_case , bos_token_id=snake_case , eos_token_id=snake_case , **snake_case )
snake_case_ :Optional[int] = vocab_size
snake_case_ :Optional[int] = max_position_embeddings
snake_case_ :Union[str, Any] = hidden_size
snake_case_ :Optional[int] = visual_embedding_dim
snake_case_ :int = num_hidden_layers
snake_case_ :Optional[int] = num_attention_heads
snake_case_ :Optional[int] = intermediate_size
snake_case_ :str = hidden_act
snake_case_ :Optional[Any] = hidden_dropout_prob
snake_case_ :str = attention_probs_dropout_prob
snake_case_ :List[Any] = initializer_range
snake_case_ :Optional[Any] = type_vocab_size
snake_case_ :Tuple = layer_norm_eps
snake_case_ :Optional[Any] = bypass_transformer
snake_case_ :List[str] = special_visual_initialize
| 66 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
print(f"""Vertex\tShortest Distance from vertex {src}""" )
for i, d in enumerate(_SCREAMING_SNAKE_CASE ):
print(f"""{i}\t\t{d}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
return True
return False
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> list[float]:
snake_case_ = [float("""inf""" )] * vertex_count
snake_case_ = 0.0
for _ in range(vertex_count - 1 ):
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
snake_case_ = distance[u] + w
snake_case_ = check_negative_cycle(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if negative_cycle_exists:
raise Exception("""Negative cycle found""" )
return distance
if __name__ == "__main__":
import doctest
doctest.testmod()
__SCREAMING_SNAKE_CASE : int = int(input('Enter number of vertices: ').strip())
__SCREAMING_SNAKE_CASE : Dict = int(input('Enter number of edges: ').strip())
__SCREAMING_SNAKE_CASE : list[dict[str, int]] = [{} for _ in range(E)]
for i in range(E):
print('Edge ', i + 1)
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE : List[str] = (
int(x)
for x in input('Enter source, destination, weight: ').strip().split(' ')
)
__SCREAMING_SNAKE_CASE : Union[str, Any] = {'src': src, 'dst': dest, 'weight': weight}
__SCREAMING_SNAKE_CASE : Union[str, Any] = int(input('\nEnter shortest path source:').strip())
__SCREAMING_SNAKE_CASE : str = bellman_ford(graph, V, E, source)
print_distance(shortest_distance, 0)
| 347 | 0 |
'''simple docstring'''
import contextlib
import csv
import json
import os
import sqlitea
import tarfile
import textwrap
import zipfile
import pyarrow as pa
import pyarrow.parquet as pq
import pytest
import datasets
import datasets.config
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( ) -> str:
__lowerCamelCase = 10
__lowerCamelCase = datasets.Features(
{
'''tokens''': datasets.Sequence(datasets.Value('''string''' ) ),
'''labels''': datasets.Sequence(datasets.ClassLabel(names=['''negative''', '''positive'''] ) ),
'''answers''': datasets.Sequence(
{
'''text''': datasets.Value('''string''' ),
'''answer_start''': datasets.Value('''int32''' ),
} ),
'''id''': datasets.Value('''int64''' ),
} )
__lowerCamelCase = datasets.Dataset.from_dict(
{
'''tokens''': [['''foo'''] * 5] * n,
'''labels''': [[1] * 5] * n,
'''answers''': [{'''answer_start''': [97], '''text''': ['''1976''']}] * 10,
'''id''': list(range(UpperCamelCase__ ) ),
} , features=UpperCamelCase__ , )
return dataset
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> List[Any]:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''file.arrow''' )
dataset.map(cache_file_name=UpperCamelCase__ )
return filename
# FILE_CONTENT + files
__UpperCAmelCase ="\\n Text data.\n Second line of data."
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Union[str, Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt'''
__lowerCamelCase = FILE_CONTENT
with open(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ )
return filename
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Optional[Any]:
import bza
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.bz2'''
__lowerCamelCase = bytes(UpperCamelCase__ , '''utf-8''' )
with bza.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Dict:
import gzip
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''file.txt.gz''' )
__lowerCamelCase = bytes(UpperCamelCase__ , '''utf-8''' )
with gzip.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> str:
if datasets.config.LZ4_AVAILABLE:
import lza.frame
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.lz4'''
__lowerCamelCase = bytes(UpperCamelCase__ , '''utf-8''' )
with lza.frame.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> int:
if datasets.config.PY7ZR_AVAILABLE:
import pyazr
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.7z'''
with pyazr.SevenZipFile(UpperCamelCase__ , '''w''' ) as archive:
archive.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> List[Any]:
import tarfile
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.tar'''
with tarfile.TarFile(UpperCamelCase__ , '''w''' ) as f:
f.add(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> str:
import lzma
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.xz'''
__lowerCamelCase = bytes(UpperCamelCase__ , '''utf-8''' )
with lzma.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> Optional[Any]:
import zipfile
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> List[Any]:
if datasets.config.ZSTANDARD_AVAILABLE:
import zstandard as zstd
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.txt.zst'''
__lowerCamelCase = bytes(UpperCamelCase__ , '''utf-8''' )
with zstd.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Tuple:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''file.xml'''
__lowerCamelCase = textwrap.dedent(
'''\
<?xml version="1.0" encoding="UTF-8" ?>
<tmx version="1.4">
<header segtype="sentence" srclang="ca" />
<body>
<tu>
<tuv xml:lang="ca"><seg>Contingut 1</seg></tuv>
<tuv xml:lang="en"><seg>Content 1</seg></tuv>
</tu>
<tu>
<tuv xml:lang="ca"><seg>Contingut 2</seg></tuv>
<tuv xml:lang="en"><seg>Content 2</seg></tuv>
</tu>
<tu>
<tuv xml:lang="ca"><seg>Contingut 3</seg></tuv>
<tuv xml:lang="en"><seg>Content 3</seg></tuv>
</tu>
<tu>
<tuv xml:lang="ca"><seg>Contingut 4</seg></tuv>
<tuv xml:lang="en"><seg>Content 4</seg></tuv>
</tu>
<tu>
<tuv xml:lang="ca"><seg>Contingut 5</seg></tuv>
<tuv xml:lang="en"><seg>Content 5</seg></tuv>
</tu>
</body>
</tmx>''' )
with open(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ )
return filename
__UpperCAmelCase =[
{"col_1": "0", "col_2": 0, "col_3": 0.0},
{"col_1": "1", "col_2": 1, "col_3": 1.0},
{"col_1": "2", "col_2": 2, "col_3": 2.0},
{"col_1": "3", "col_2": 3, "col_3": 3.0},
]
__UpperCAmelCase =[
{"col_1": "4", "col_2": 4, "col_3": 4.0},
{"col_1": "5", "col_2": 5, "col_3": 5.0},
]
__UpperCAmelCase ={
"col_1": ["0", "1", "2", "3"],
"col_2": [0, 1, 2, 3],
"col_3": [0.0, 1.0, 2.0, 3.0],
}
__UpperCAmelCase =[
{"col_3": 0.0, "col_1": "0", "col_2": 0},
{"col_3": 1.0, "col_1": "1", "col_2": 1},
]
__UpperCAmelCase =[
{"col_1": "s0", "col_2": 0, "col_3": 0.0},
{"col_1": "s1", "col_2": 1, "col_3": 1.0},
{"col_1": "s2", "col_2": 2, "col_3": 2.0},
{"col_1": "s3", "col_2": 3, "col_3": 3.0},
]
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( ) -> Any:
return DATA_DICT_OF_LISTS
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> List[str]:
__lowerCamelCase = datasets.Dataset.from_dict(UpperCamelCase__ )
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.arrow''' )
dataset.map(cache_file_name=UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> int:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.sqlite''' )
with contextlib.closing(sqlitea.connect(UpperCamelCase__ ) ) as con:
__lowerCamelCase = con.cursor()
cur.execute('''CREATE TABLE dataset(col_1 text, col_2 int, col_3 real)''' )
for item in DATA:
cur.execute('''INSERT INTO dataset(col_1, col_2, col_3) VALUES (?, ?, ?)''' , tuple(item.values() ) )
con.commit()
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Any:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.csv''' )
with open(UpperCamelCase__ , '''w''' , newline='''''' ) as f:
__lowerCamelCase = csv.DictWriter(UpperCamelCase__ , fieldnames=['''col_1''', '''col_2''', '''col_3'''] )
writer.writeheader()
for item in DATA:
writer.writerow(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Dict:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset2.csv''' )
with open(UpperCamelCase__ , '''w''' , newline='''''' ) as f:
__lowerCamelCase = csv.DictWriter(UpperCamelCase__ , fieldnames=['''col_1''', '''col_2''', '''col_3'''] )
writer.writeheader()
for item in DATA:
writer.writerow(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> List[Any]:
import bza
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.csv.bz2'''
with open(UpperCamelCase__ , '''rb''' ) as f:
__lowerCamelCase = f.read()
# data = bytes(FILE_CONTENT, "utf-8")
with bza.open(UpperCamelCase__ , '''wb''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> int:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.csv.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> List[str]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.csv.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(csv_path.replace('''.csv''' , '''.CSV''' ) ) )
f.write(UpperCamelCase__ , arcname=os.path.basename(csva_path.replace('''.csv''' , '''.CSV''' ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> Any:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset_with_dir.csv.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Union[str, Any]:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.parquet''' )
__lowerCamelCase = pa.schema(
{
'''col_1''': pa.string(),
'''col_2''': pa.intaa(),
'''col_3''': pa.floataa(),
} )
with open(UpperCamelCase__ , '''wb''' ) as f:
__lowerCamelCase = pq.ParquetWriter(UpperCamelCase__ , schema=UpperCamelCase__ )
__lowerCamelCase = pa.Table.from_pydict({k: [DATA[i][k] for i in range(len(UpperCamelCase__ ) )] for k in DATA[0]} , schema=UpperCamelCase__ )
writer.write_table(UpperCamelCase__ )
writer.close()
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Union[str, Any]:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.json''' )
__lowerCamelCase = {'''data''': DATA}
with open(UpperCamelCase__ , '''w''' ) as f:
json.dump(UpperCamelCase__ , UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Tuple:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.json''' )
__lowerCamelCase = {'''data''': DATA_DICT_OF_LISTS}
with open(UpperCamelCase__ , '''w''' ) as f:
json.dump(UpperCamelCase__ , UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Dict:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.jsonl''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in DATA:
f.write(json.dumps(UpperCamelCase__ ) + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> str:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset2.jsonl''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in DATA:
f.write(json.dumps(UpperCamelCase__ ) + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Dict:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset_312.jsonl''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in DATA_312:
f.write(json.dumps(UpperCamelCase__ ) + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Optional[int]:
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset-str.jsonl''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in DATA_STR:
f.write(json.dumps(UpperCamelCase__ ) + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> str:
import gzip
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.txt.gz''' )
with open(UpperCamelCase__ , '''rb''' ) as orig_file:
with gzip.open(UpperCamelCase__ , '''wb''' ) as zipped_file:
zipped_file.writelines(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> int:
import gzip
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.jsonl.gz''' )
with open(UpperCamelCase__ , '''rb''' ) as orig_file:
with gzip.open(UpperCamelCase__ , '''wb''' ) as zipped_file:
zipped_file.writelines(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> List[str]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.jsonl.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> List[Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset_nested.jsonl.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.join('''nested''' , os.path.basename(UpperCamelCase__ ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> str:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset_with_dir.jsonl.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> Any:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.jsonl.tar'''
with tarfile.TarFile(UpperCamelCase__ , '''w''' ) as f:
f.add(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
f.add(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> Optional[Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset_nested.jsonl.tar'''
with tarfile.TarFile(UpperCamelCase__ , '''w''' ) as f:
f.add(UpperCamelCase__ , arcname=os.path.join('''nested''' , os.path.basename(UpperCamelCase__ ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Any:
__lowerCamelCase = ['''0''', '''1''', '''2''', '''3''']
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset.txt''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in data:
f.write(item + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> Optional[Any]:
__lowerCamelCase = ['''0''', '''1''', '''2''', '''3''']
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset2.txt''' )
with open(UpperCamelCase__ , '''w''' ) as f:
for item in data:
f.write(item + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> List[Any]:
__lowerCamelCase = ['''0''', '''1''', '''2''', '''3''']
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.abc'''
with open(UpperCamelCase__ , '''w''' ) as f:
for item in data:
f.write(item + '''\n''' )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> Union[str, Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.text.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> str:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset_with_dir.text.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
f.write(UpperCamelCase__ , arcname=os.path.join('''main_dir''' , os.path.basename(UpperCamelCase__ ) ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ ) -> Union[str, Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.ext.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename('''unsupported.ext''' ) )
f.write(UpperCamelCase__ , arcname=os.path.basename('''unsupported_2.ext''' ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> List[str]:
__lowerCamelCase = '''\n'''.join(['''First''', '''Second\u2029with Unicode new line''', '''Third'''] )
__lowerCamelCase = str(tmp_path_factory.mktemp('''data''' ) / '''dataset_with_unicode_new_lines.txt''' )
with open(UpperCamelCase__ , '''w''' , encoding='''utf-8''' ) as f:
f.write(UpperCamelCase__ )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( ) -> Optional[Any]:
return os.path.join('''tests''' , '''features''' , '''data''' , '''test_image_rgb.jpg''' )
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( ) -> Optional[Any]:
return os.path.join('''tests''' , '''features''' , '''data''' , '''test_audio_44100.wav''' )
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ , UpperCamelCase__ ) -> Optional[Any]:
__lowerCamelCase = tmp_path_factory.mktemp('''data''' ) / '''dataset.img.zip'''
with zipfile.ZipFile(UpperCamelCase__ , '''w''' ) as f:
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ) )
f.write(UpperCamelCase__ , arcname=os.path.basename(UpperCamelCase__ ).replace('''.jpg''' , '''2.jpg''' ) )
return path
@pytest.fixture(scope='''session''' )
def __lowerCAmelCase ( UpperCamelCase__ ) -> int:
__lowerCamelCase = tmp_path_factory.mktemp('''data_dir''' )
(data_dir / "subdir").mkdir()
with open(data_dir / '''subdir''' / '''train.txt''' , '''w''' ) as f:
f.write('''foo\n''' * 10 )
with open(data_dir / '''subdir''' / '''test.txt''' , '''w''' ) as f:
f.write('''bar\n''' * 10 )
# hidden file
with open(data_dir / '''subdir''' / '''.test.txt''' , '''w''' ) as f:
f.write('''bar\n''' * 10 )
# hidden directory
(data_dir / ".subdir").mkdir()
with open(data_dir / '''.subdir''' / '''train.txt''' , '''w''' ) as f:
f.write('''foo\n''' * 10 )
with open(data_dir / '''.subdir''' / '''test.txt''' , '''w''' ) as f:
f.write('''bar\n''' * 10 )
return data_dir
| 67 |
"""simple docstring"""
import argparse
import logging
import os
import re
import tensorflow as tf
from transformers import (
AutoConfig,
AutoTokenizer,
DataCollatorForLanguageModeling,
PushToHubCallback,
TFAutoModelForMaskedLM,
create_optimizer,
)
__SCREAMING_SNAKE_CASE : List[str] = logging.getLogger(__name__)
__SCREAMING_SNAKE_CASE : str = tf.data.AUTOTUNE
def _a ( ) -> List[str]:
snake_case_ = argparse.ArgumentParser(description="""Train a masked language model on TPU.""" )
parser.add_argument(
"""--pretrained_model_config""" , type=_SCREAMING_SNAKE_CASE , default="""roberta-base""" , help="""The model config to use. Note that we don't copy the model's weights, only the config!""" , )
parser.add_argument(
"""--tokenizer""" , type=_SCREAMING_SNAKE_CASE , default="""unigram-tokenizer-wikitext""" , help="""The name of the tokenizer to load. We use the pretrained tokenizer to initialize the model's vocab size.""" , )
parser.add_argument(
"""--per_replica_batch_size""" , type=_SCREAMING_SNAKE_CASE , default=8 , help="""Batch size per TPU core.""" , )
parser.add_argument(
"""--no_tpu""" , action="""store_true""" , help="""If set, run on CPU and don't try to initialize a TPU. Useful for debugging on non-TPU instances.""" , )
parser.add_argument(
"""--tpu_name""" , type=_SCREAMING_SNAKE_CASE , help="""Name of TPU resource to initialize. Should be blank on Colab, and 'local' on TPU VMs.""" , default="""local""" , )
parser.add_argument(
"""--tpu_zone""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud zone that TPU resource is located in. Only used for non-Colab TPU nodes.""" , )
parser.add_argument(
"""--gcp_project""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud project name. Only used for non-Colab TPU nodes.""" )
parser.add_argument(
"""--bfloat16""" , action="""store_true""" , help="""Use mixed-precision bfloat16 for training. This is the recommended lower-precision format for TPU.""" , )
parser.add_argument(
"""--train_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to training dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--shuffle_buffer_size""" , type=_SCREAMING_SNAKE_CASE , default=2**18 , help="""Size of the shuffle buffer (in samples)""" , )
parser.add_argument(
"""--eval_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to evaluation dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--num_epochs""" , type=_SCREAMING_SNAKE_CASE , default=1 , help="""Number of epochs to train for.""" , )
parser.add_argument(
"""--learning_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-4 , help="""Learning rate to use for training.""" , )
parser.add_argument(
"""--weight_decay_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-3 , help="""Weight decay rate to use for training.""" , )
parser.add_argument(
"""--max_length""" , type=_SCREAMING_SNAKE_CASE , default=512 , help="""Maximum length of tokenized sequences. Should match the setting used in prepare_tfrecord_shards.py""" , )
parser.add_argument(
"""--mlm_probability""" , type=_SCREAMING_SNAKE_CASE , default=0.15 , help="""Fraction of tokens to mask during training.""" , )
parser.add_argument("""--output_dir""" , type=_SCREAMING_SNAKE_CASE , required=_SCREAMING_SNAKE_CASE , help="""Path to save model checkpoints to.""" )
parser.add_argument("""--hub_model_id""" , type=_SCREAMING_SNAKE_CASE , help="""Model ID to upload to on the Hugging Face Hub.""" )
snake_case_ = parser.parse_args()
return args
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
try:
if args.tpu_name:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver(
args.tpu_name , zone=args.tpu_zone , project=args.gcp_project )
else:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
raise RuntimeError(
"""Couldn't connect to TPU! Most likely you need to specify --tpu_name, --tpu_zone, or """
"""--gcp_project. When running on a TPU VM, use --tpu_name local.""" )
tf.config.experimental_connect_to_cluster(_SCREAMING_SNAKE_CASE )
tf.tpu.experimental.initialize_tpu_system(_SCREAMING_SNAKE_CASE )
return tpu
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = 0
for file in file_list:
snake_case_ = file.split("""/""" )[-1]
snake_case_ = re.search(r"""-\d+-(\d+)\.tfrecord""" , _SCREAMING_SNAKE_CASE ).group(1 )
snake_case_ = int(_SCREAMING_SNAKE_CASE )
num_samples += sample_count
return num_samples
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None ) -> Union[str, Any]:
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.data.Dataset.from_tensor_slices(_SCREAMING_SNAKE_CASE )
if shuffle:
snake_case_ = dataset.shuffle(len(_SCREAMING_SNAKE_CASE ) )
snake_case_ = tf.data.TFRecordDataset(_SCREAMING_SNAKE_CASE , num_parallel_reads=_SCREAMING_SNAKE_CASE )
# TF can't infer the total sample count because it doesn't read all the records yet, so we assert it here
snake_case_ = dataset.apply(tf.data.experimental.assert_cardinality(_SCREAMING_SNAKE_CASE ) )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
if shuffle:
assert shuffle_buffer_size is not None
snake_case_ = dataset.shuffle(args.shuffle_buffer_size )
snake_case_ = dataset.batch(_SCREAMING_SNAKE_CASE , drop_remainder=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.prefetch(_SCREAMING_SNAKE_CASE )
return dataset
def _a ( _SCREAMING_SNAKE_CASE ) -> List[Any]:
if not args.no_tpu:
snake_case_ = initialize_tpu(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.distribute.TPUStrategy(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = tf.distribute.OneDeviceStrategy(device="""/gpu:0""" )
if args.bfloataa:
tf.keras.mixed_precision.set_global_policy("""mixed_bfloat16""" )
snake_case_ = AutoTokenizer.from_pretrained(args.tokenizer )
snake_case_ = AutoConfig.from_pretrained(args.pretrained_model_config )
snake_case_ = tokenizer.vocab_size
snake_case_ = tf.io.gfile.glob(os.path.join(args.train_dataset , """*.tfrecord""" ) )
if not training_records:
raise ValueError(f"""No .tfrecord files found in {args.train_dataset}.""" )
snake_case_ = tf.io.gfile.glob(os.path.join(args.eval_dataset , """*.tfrecord""" ) )
if not eval_records:
raise ValueError(f"""No .tfrecord files found in {args.eval_dataset}.""" )
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = num_train_samples // (args.per_replica_batch_size * strategy.num_replicas_in_sync)
snake_case_ = steps_per_epoch * args.num_epochs
with strategy.scope():
snake_case_ = TFAutoModelForMaskedLM.from_config(_SCREAMING_SNAKE_CASE )
model(model.dummy_inputs ) # Pass some dummy inputs through the model to ensure all the weights are built
snake_case_ , snake_case_ = create_optimizer(
num_train_steps=_SCREAMING_SNAKE_CASE , num_warmup_steps=total_train_steps // 20 , init_lr=args.learning_rate , weight_decay_rate=args.weight_decay_rate , )
# Transformers models compute the right loss for their task by default when labels are passed, and will
# use this for training unless you specify your own loss function in compile().
model.compile(optimizer=_SCREAMING_SNAKE_CASE , metrics=["""accuracy"""] )
def decode_fn(_SCREAMING_SNAKE_CASE ):
snake_case_ = {
"""input_ids""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
"""attention_mask""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
}
return tf.io.parse_single_example(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# Many of the data collators in Transformers are TF-compilable when return_tensors == "tf", so we can
# use their methods in our data pipeline.
snake_case_ = DataCollatorForLanguageModeling(
tokenizer=_SCREAMING_SNAKE_CASE , mlm_probability=args.mlm_probability , mlm=_SCREAMING_SNAKE_CASE , return_tensors="""tf""" )
def mask_with_collator(_SCREAMING_SNAKE_CASE ):
# TF really needs an isin() function
snake_case_ = (
~tf.cast(batch["""attention_mask"""] , tf.bool )
| (batch["""input_ids"""] == tokenizer.cls_token_id)
| (batch["""input_ids"""] == tokenizer.sep_token_id)
)
snake_case_ , snake_case_ = data_collator.tf_mask_tokens(
batch["""input_ids"""] , vocab_size=len(_SCREAMING_SNAKE_CASE ) , mask_token_id=tokenizer.mask_token_id , special_tokens_mask=_SCREAMING_SNAKE_CASE , )
return batch
snake_case_ = args.per_replica_batch_size * strategy.num_replicas_in_sync
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , shuffle_buffer_size=args.shuffle_buffer_size , )
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , )
snake_case_ = []
if args.hub_model_id:
callbacks.append(
PushToHubCallback(output_dir=args.output_dir , hub_model_id=args.hub_model_id , tokenizer=_SCREAMING_SNAKE_CASE ) )
model.fit(
_SCREAMING_SNAKE_CASE , validation_data=_SCREAMING_SNAKE_CASE , epochs=args.num_epochs , callbacks=_SCREAMING_SNAKE_CASE , )
model.save_pretrained(args.output_dir )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : Union[str, Any] = parse_args()
main(args)
| 347 | 0 |
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
lowerCAmelCase__ = {
"""configuration_instructblip""": [
"""INSTRUCTBLIP_PRETRAINED_CONFIG_ARCHIVE_MAP""",
"""InstructBlipConfig""",
"""InstructBlipQFormerConfig""",
"""InstructBlipVisionConfig""",
],
"""processing_instructblip""": ["""InstructBlipProcessor"""],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowerCAmelCase__ = [
"""INSTRUCTBLIP_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""InstructBlipQFormerModel""",
"""InstructBlipPreTrainedModel""",
"""InstructBlipForConditionalGeneration""",
"""InstructBlipVisionModel""",
]
if TYPE_CHECKING:
from .configuration_instructblip import (
INSTRUCTBLIP_PRETRAINED_CONFIG_ARCHIVE_MAP,
InstructBlipConfig,
InstructBlipQFormerConfig,
InstructBlipVisionConfig,
)
from .processing_instructblip import InstructBlipProcessor
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_instructblip import (
INSTRUCTBLIP_PRETRAINED_MODEL_ARCHIVE_LIST,
InstructBlipForConditionalGeneration,
InstructBlipPreTrainedModel,
InstructBlipQFormerModel,
InstructBlipVisionModel,
)
else:
import sys
lowerCAmelCase__ = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__)
| 68 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> float:
if density <= 0:
raise ValueError("""Impossible fluid density""" )
if bulk_modulus <= 0:
raise ValueError("""Impossible bulk modulus""" )
return (bulk_modulus / density) ** 0.5
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
"""simple docstring"""
import json
import logging
import math
import os
import sys
from dataclasses import dataclass, field
from typing import Optional
from datasets import Dataset, load_dataset
import transformers
from transformers import (
CONFIG_MAPPING,
MODEL_FOR_MASKED_LM_MAPPING,
AutoConfig,
AutoModelForMaskedLM,
AutoTokenizer,
DataCollatorForWholeWordMask,
HfArgumentParser,
Trainer,
TrainingArguments,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint, is_main_process
__UpperCamelCase = logging.getLogger(__name__)
__UpperCamelCase = list(MODEL_FOR_MASKED_LM_MAPPING.keys())
__UpperCamelCase = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)
@dataclass
class UpperCamelCase :
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={
"help": (
"The model checkpoint for weights initialization.Don't set if you want to train a model from scratch."
)
} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "If training from scratch, pass a model type from the list: " + ", ".join(lowerCAmelCase__ )} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={
"help": (
"Override some existing default config settings when a model is trained from scratch. Example: "
"n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index"
)
} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "Pretrained config name or path if not the same as model_name"} )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."} , )
SCREAMING_SNAKE_CASE_ = field(
default="main" , metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={
"help": (
"Will use the token generated when running `huggingface-cli login` (necessary to use this script "
"with private models)."
)
} , )
def a_ ( self) -> Tuple:
if self.config_overrides is not None and (self.config_name is not None or self.model_name_or_path is not None):
raise ValueError(
'--config_overrides can\'t be used in combination with --config_name or --model_name_or_path')
@dataclass
class UpperCamelCase :
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "The name of the dataset to use (via the datasets library)."} )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} )
SCREAMING_SNAKE_CASE_ = field(default=lowerCAmelCase__ , metadata={"help": "The input training data file (a text file)."} )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "An optional input evaluation data file to evaluate the perplexity on (a text file)."} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "An optional input train ref data file for whole word masking in Chinese."} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "An optional input validation ref data file for whole word masking in Chinese."} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "Overwrite the cached training and evaluation sets"} )
SCREAMING_SNAKE_CASE_ = field(
default=5 , metadata={
"help": "The percentage of the train set used as validation set in case there's no validation split"
} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={
"help": (
"The maximum total input sequence length after tokenization. Sequences longer "
"than this will be truncated. Default to the max input length of the model."
)
} , )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={"help": "The number of processes to use for the preprocessing."} , )
SCREAMING_SNAKE_CASE_ = field(
default=0.15 , metadata={"help": "Ratio of tokens to mask for masked language modeling loss"} )
SCREAMING_SNAKE_CASE_ = field(
default=lowerCAmelCase__ , metadata={
"help": (
"Whether to pad all samples to `max_seq_length`. "
"If False, will pad the samples dynamically when batching to the maximum length in the batch."
)
} , )
def a_ ( self) -> Optional[Any]:
if self.train_file is not None:
snake_case_ = self.train_file.split('.')[-1]
assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, a json or a txt file."
if self.validation_file is not None:
snake_case_ = self.validation_file.split('.')[-1]
assert extension in ["csv", "json", "txt"], "`validation_file` should be a csv, a json or a txt file."
def UpperCAmelCase ( UpperCAmelCase , UpperCAmelCase ) -> str:
with open(UpperCAmelCase , 'r' , encoding='utf-8' ) as f:
snake_case_ = [json.loads(UpperCAmelCase ) for line in f.read().splitlines() if (len(UpperCAmelCase ) > 0 and not line.isspace())]
assert len(UpperCAmelCase ) == len(UpperCAmelCase )
snake_case_ = {c: dataset[c] for c in dataset.column_names}
snake_case_ = refs
return Dataset.from_dict(UpperCAmelCase )
def UpperCAmelCase ( ) -> Optional[Any]:
# See all possible arguments in src/transformers/training_args.py
# or by passing the --help flag to this script.
# We now keep distinct sets of args, for a cleaner separation of concerns.
snake_case_ = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments) )
if len(sys.argv ) == 2 and sys.argv[1].endswith('.json' ):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
snake_case_ , snake_case_ , snake_case_ = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1] ) )
else:
snake_case_ , snake_case_ , snake_case_ = parser.parse_args_into_dataclasses()
# Detecting last checkpoint.
snake_case_ = None
if os.path.isdir(training_args.output_dir ) and training_args.do_train and not training_args.overwrite_output_dir:
snake_case_ = get_last_checkpoint(training_args.output_dir )
if last_checkpoint is None and len(os.listdir(training_args.output_dir ) ) > 0:
raise ValueError(
f'Output directory ({training_args.output_dir}) already exists and is not empty. '
'Use --overwrite_output_dir to overcome.' )
elif last_checkpoint is not None:
logger.info(
f'Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change '
'the `--output_dir` or add `--overwrite_output_dir` to train from scratch.' )
# Setup logging
logging.basicConfig(
format='%(asctime)s - %(levelname)s - %(name)s - %(message)s' , datefmt='%m/%d/%Y %H:%M:%S' , handlers=[logging.StreamHandler(sys.stdout )] , )
logger.setLevel(logging.INFO if is_main_process(training_args.local_rank ) else logging.WARN )
# Log on each process the small summary:
logger.warning(
f'Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}'
+ f'distributed training: {bool(training_args.local_rank != -1 )}, 16-bits training: {training_args.fpaa}' )
# Set the verbosity to info of the Transformers logger (on main process only):
if is_main_process(training_args.local_rank ):
transformers.utils.logging.set_verbosity_info()
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
logger.info('Training/evaluation parameters %s' , UpperCAmelCase )
# Set seed before initializing model.
set_seed(training_args.seed )
# Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below)
# or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/
# (the dataset will be downloaded automatically from the datasets Hub).
#
# For CSV/JSON files, this script will use the column called 'text' or the first column if no column called
# 'text' is found. You can easily tweak this behavior (see below).
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
snake_case_ = load_dataset(data_args.dataset_name , data_args.dataset_config_name )
if "validation" not in datasets.keys():
snake_case_ = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , split=f'train[:{data_args.validation_split_percentage}%]' , )
snake_case_ = load_dataset(
data_args.dataset_name , data_args.dataset_config_name , split=f'train[{data_args.validation_split_percentage}%:]' , )
else:
snake_case_ = {}
if data_args.train_file is not None:
snake_case_ = data_args.train_file
if data_args.validation_file is not None:
snake_case_ = data_args.validation_file
snake_case_ = data_args.train_file.split('.' )[-1]
if extension == "txt":
snake_case_ = 'text'
snake_case_ = load_dataset(UpperCAmelCase , data_files=UpperCAmelCase )
# See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
# https://huggingface.co/docs/datasets/loading_datasets.html.
# Load pretrained model and tokenizer
#
# Distributed training:
# The .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
snake_case_ = {
'cache_dir': model_args.cache_dir,
'revision': model_args.model_revision,
'use_auth_token': True if model_args.use_auth_token else None,
}
if model_args.config_name:
snake_case_ = AutoConfig.from_pretrained(model_args.config_name , **UpperCAmelCase )
elif model_args.model_name_or_path:
snake_case_ = AutoConfig.from_pretrained(model_args.model_name_or_path , **UpperCAmelCase )
else:
snake_case_ = CONFIG_MAPPING[model_args.model_type]()
logger.warning('You are instantiating a new config instance from scratch.' )
if model_args.config_overrides is not None:
logger.info(f'Overriding config: {model_args.config_overrides}' )
config.update_from_string(model_args.config_overrides )
logger.info(f'New config: {config}' )
snake_case_ = {
'cache_dir': model_args.cache_dir,
'use_fast': model_args.use_fast_tokenizer,
'revision': model_args.model_revision,
'use_auth_token': True if model_args.use_auth_token else None,
}
if model_args.tokenizer_name:
snake_case_ = AutoTokenizer.from_pretrained(model_args.tokenizer_name , **UpperCAmelCase )
elif model_args.model_name_or_path:
snake_case_ = AutoTokenizer.from_pretrained(model_args.model_name_or_path , **UpperCAmelCase )
else:
raise ValueError(
'You are instantiating a new tokenizer from scratch. This is not supported by this script.'
'You can do it from another script, save it, and load it from here, using --tokenizer_name.' )
if model_args.model_name_or_path:
snake_case_ = AutoModelForMaskedLM.from_pretrained(
model_args.model_name_or_path , from_tf=bool('.ckpt' in model_args.model_name_or_path ) , config=UpperCAmelCase , cache_dir=model_args.cache_dir , revision=model_args.model_revision , use_auth_token=True if model_args.use_auth_token else None , )
else:
logger.info('Training new model from scratch' )
snake_case_ = AutoModelForMaskedLM.from_config(UpperCAmelCase )
model.resize_token_embeddings(len(UpperCAmelCase ) )
# Preprocessing the datasets.
# First we tokenize all the texts.
if training_args.do_train:
snake_case_ = datasets['train'].column_names
else:
snake_case_ = datasets['validation'].column_names
snake_case_ = 'text' if 'text' in column_names else column_names[0]
snake_case_ = 'max_length' if data_args.pad_to_max_length else False
def tokenize_function(UpperCAmelCase ):
# Remove empty lines
snake_case_ = [line for line in examples['text'] if len(UpperCAmelCase ) > 0 and not line.isspace()]
return tokenizer(examples['text'] , padding=UpperCAmelCase , truncation=UpperCAmelCase , max_length=data_args.max_seq_length )
snake_case_ = datasets.map(
UpperCAmelCase , batched=UpperCAmelCase , num_proc=data_args.preprocessing_num_workers , remove_columns=[text_column_name] , load_from_cache_file=not data_args.overwrite_cache , )
# Add the chinese references if provided
if data_args.train_ref_file is not None:
snake_case_ = add_chinese_references(tokenized_datasets['train'] , data_args.train_ref_file )
if data_args.validation_ref_file is not None:
snake_case_ = add_chinese_references(
tokenized_datasets['validation'] , data_args.validation_ref_file )
# If we have ref files, need to avoid it removed by trainer
snake_case_ = data_args.train_ref_file or data_args.validation_ref_file
if has_ref:
snake_case_ = False
# Data collator
# This one will take care of randomly masking the tokens.
snake_case_ = DataCollatorForWholeWordMask(tokenizer=UpperCAmelCase , mlm_probability=data_args.mlm_probability )
# Initialize our Trainer
snake_case_ = Trainer(
model=UpperCAmelCase , args=UpperCAmelCase , train_dataset=tokenized_datasets['train'] if training_args.do_train else None , eval_dataset=tokenized_datasets['validation'] if training_args.do_eval else None , tokenizer=UpperCAmelCase , data_collator=UpperCAmelCase , )
# Training
if training_args.do_train:
if last_checkpoint is not None:
snake_case_ = last_checkpoint
elif model_args.model_name_or_path is not None and os.path.isdir(model_args.model_name_or_path ):
snake_case_ = model_args.model_name_or_path
else:
snake_case_ = None
snake_case_ = trainer.train(resume_from_checkpoint=UpperCAmelCase )
trainer.save_model() # Saves the tokenizer too for easy upload
snake_case_ = os.path.join(training_args.output_dir , 'train_results.txt' )
if trainer.is_world_process_zero():
with open(UpperCAmelCase , 'w' ) as writer:
logger.info('***** Train results *****' )
for key, value in sorted(train_result.metrics.items() ):
logger.info(f' {key} = {value}' )
writer.write(f'{key} = {value}\n' )
# Need to save the state, since Trainer.save_model saves only the tokenizer with the model
trainer.state.save_to_json(os.path.join(training_args.output_dir , 'trainer_state.json' ) )
# Evaluation
snake_case_ = {}
if training_args.do_eval:
logger.info('*** Evaluate ***' )
snake_case_ = trainer.evaluate()
snake_case_ = math.exp(eval_output['eval_loss'] )
snake_case_ = perplexity
snake_case_ = os.path.join(training_args.output_dir , 'eval_results_mlm_wwm.txt' )
if trainer.is_world_process_zero():
with open(UpperCAmelCase , 'w' ) as writer:
logger.info('***** Eval results *****' )
for key, value in sorted(results.items() ):
logger.info(f' {key} = {value}' )
writer.write(f'{key} = {value}\n' )
return results
def UpperCAmelCase ( UpperCAmelCase ) -> List[str]:
# For xla_spawn (TPUs)
main()
if __name__ == "__main__":
main()
| 69 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE ) -> bool:
if num < 0:
return False
snake_case_ = num
snake_case_ = 0
while num > 0:
snake_case_ = rev_num * 10 + (num % 10)
num //= 10
return num_copy == rev_num
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
'''simple docstring'''
# XXX: we want transformers master here - in the absense of conftest manipulating sys.path:
# hack it in for now:
import sys
from pathlib import Path
A__ : Dict =Path(__file__).resolve().parents[3] / '''src'''
sys.path.insert(1, str(git_repo_path))
import dataclasses # noqa
import io # noqa
import itertools # noqa
import json # noqa
import os # noqa
import unittest # noqa
from copy import deepcopy # noqa
from parameterized import parameterized # noqa
from transformers import TrainingArguments, is_torch_available # noqa
from transformers.deepspeed import is_deepspeed_available # noqa
from transformers.file_utils import WEIGHTS_NAME # noqa
from transformers.testing_utils import ( # noqa
CaptureLogger,
ExtendSysPath,
TestCasePlus,
execute_subprocess_async,
get_gpu_count,
mockenv_context,
require_deepspeed,
require_torch_gpu,
require_torch_multi_gpu,
slow,
)
from transformers.trainer_utils import set_seed # noqa
set_seed(42)
A__ : Optional[int] ={'''base''': '''patrickvonplaten/wav2vec2_tiny_random''', '''robust''': '''patrickvonplaten/wav2vec2_tiny_random_robust'''}
A__ : List[str] ='''zero2'''
A__ : List[Any] ='''zero3'''
A__ : str =[ZEROa, ZEROa]
def UpperCamelCase__ ( lowerCAmelCase , lowerCAmelCase , lowerCAmelCase ):
"""simple docstring"""
_lowerCAmelCase = parameterized.to_safe_name("""_""".join(str(lowerCAmelCase ) for x in param.args ) )
return f"{func.__name__}_{param_based_name}"
# Cartesian-product of zero stages with models to test
A__ : Any =list(itertools.product(stages, models.keys()))
@slow
@require_deepspeed
@require_torch_gpu
class UpperCAmelCase ( snake_case_ ):
@parameterized.expand(__snake_case , name_func=__snake_case )
def lowercase__ ( self : List[str] , __snake_case : Tuple , __snake_case : Optional[int] ) -> List[str]:
self.run_and_check(
stage=__snake_case , model=__snake_case , distributed=__snake_case , fpaa=__snake_case , )
@require_torch_multi_gpu
@parameterized.expand(__snake_case , name_func=__snake_case )
def lowercase__ ( self : List[Any] , __snake_case : Union[str, Any] , __snake_case : int ) -> int:
self.run_and_check(
stage=__snake_case , model=__snake_case , distributed=__snake_case , fpaa=__snake_case , )
@parameterized.expand(__snake_case , name_func=__snake_case )
def lowercase__ ( self : Any , __snake_case : Optional[int] , __snake_case : Dict ) -> Tuple:
self.run_and_check(
stage=__snake_case , model=__snake_case , distributed=__snake_case , fpaa=__snake_case , )
@require_torch_multi_gpu
@parameterized.expand(__snake_case , name_func=__snake_case )
def lowercase__ ( self : Tuple , __snake_case : Optional[int] , __snake_case : Dict ) -> Union[str, Any]:
self.run_and_check(
stage=__snake_case , model=__snake_case , distributed=__snake_case , fpaa=__snake_case , )
def lowercase__ ( self : Optional[Any] , __snake_case : Union[str, Any] ) -> Union[str, Any]:
# XXX: run_asr is premature and doesn't save any results
# so all we check for now is that the process didn't fail
pass
def lowercase__ ( self : Dict , __snake_case : str , __snake_case : str , __snake_case : int = 10 , __snake_case : bool = True , __snake_case : bool = True , __snake_case : bool = True , ) -> Optional[int]:
_lowerCAmelCase = models[model]
_lowerCAmelCase = self.run_trainer(
stage=__snake_case , model_name=__snake_case , eval_steps=__snake_case , num_train_epochs=1 , distributed=__snake_case , fpaa=__snake_case , )
self.do_checks(__snake_case )
return output_dir
def lowercase__ ( self : Optional[int] , __snake_case : str , __snake_case : str , __snake_case : int = 10 , __snake_case : int = 1 , __snake_case : bool = True , __snake_case : bool = True , ) -> Optional[Any]:
_lowerCAmelCase = self.get_auto_remove_tmp_dir("""./xxx""" , after=__snake_case )
_lowerCAmelCase = f"\n --model_name_or_path {model_name}\n --dataset_name hf-internal-testing/librispeech_asr_dummy\n --dataset_config_name clean\n --train_split_name validation\n --validation_split_name validation\n --output_dir {output_dir}\n --num_train_epochs {str(__snake_case )}\n --per_device_train_batch_size 2\n --per_device_eval_batch_size 2\n --evaluation_strategy steps\n --learning_rate 5e-4\n --warmup_steps 8\n --orthography timit\n --preprocessing_num_workers 1\n --group_by_length\n --freeze_feature_extractor\n --report_to none\n --save_steps 0\n --eval_steps {eval_steps}\n --report_to none\n ".split()
if fpaa:
args.extend(["""--fp16"""] )
# currently ds_config_wav2vec2_zero.json requires "zero_optimization.find_unused_parameters": true,
# hence the separate config files
_lowerCAmelCase = f"--deepspeed {self.test_file_dir_str}/ds_config_wav2vec2_{stage}.json".split()
_lowerCAmelCase = [f"{self.examples_dir_str}/research_projects/wav2vec2/run_asr.py"]
_lowerCAmelCase = self.get_launcher(__snake_case )
_lowerCAmelCase = launcher + script + args + ds_args
# keep for quick debug
# print(" ".join([f"\nPYTHONPATH={self.src_dir_str}"] +cmd)); die
execute_subprocess_async(__snake_case , env=self.get_env() )
return output_dir
def lowercase__ ( self : Tuple , __snake_case : List[Any]=False ) -> List[Any]:
# 1. explicitly set --num_nodes=1 just in case these tests end up run on a multi-node setup
# - it won't be able to handle that
# 2. for now testing with just 2 gpus max (since some quality tests may give different
# results with mode gpus because we use very little data)
_lowerCAmelCase = min(2 , get_gpu_count() ) if distributed else 1
return f"deepspeed --num_nodes 1 --num_gpus {num_gpus}".split()
| 70 |
"""simple docstring"""
import unittest
from transformers import SPIECE_UNDERLINE
from transformers.models.speechta import SpeechTaTokenizer
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.tokenization_utils import AddedToken
from ...test_tokenization_common import TokenizerTesterMixin
__SCREAMING_SNAKE_CASE : Tuple = get_tests_dir('fixtures/test_sentencepiece_bpe_char.model')
@require_sentencepiece
@require_tokenizers
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Tuple = SpeechTaTokenizer
__lowercase: int = False
__lowercase: List[str] = True
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
snake_case_ = SpeechTaTokenizer(UpperCAmelCase_ )
snake_case_ = AddedToken("""<mask>""" , lstrip=UpperCAmelCase_ , rstrip=UpperCAmelCase_ )
snake_case_ = mask_token
tokenizer.add_special_tokens({"""mask_token""": mask_token} )
tokenizer.add_tokens(["""<ctc_blank>"""] )
tokenizer.save_pretrained(self.tmpdirname )
def lowerCAmelCase ( self : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = """this is a test"""
snake_case_ = """this is a test"""
return input_text, output_text
def lowerCAmelCase ( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : Any=False , UpperCAmelCase_ : Tuple=20 , UpperCAmelCase_ : Dict=5 ) ->List[Any]:
"""simple docstring"""
snake_case_ , snake_case_ = self.get_input_output_texts(UpperCAmelCase_ )
snake_case_ = tokenizer.encode(UpperCAmelCase_ , add_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.decode(UpperCAmelCase_ , clean_up_tokenization_spaces=UpperCAmelCase_ )
return text, ids
def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = """<pad>"""
snake_case_ = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(UpperCAmelCase_ ) , UpperCAmelCase_ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(UpperCAmelCase_ ) , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-4] , """œ""" )
self.assertEqual(vocab_keys[-2] , """<mask>""" )
self.assertEqual(vocab_keys[-1] , """<ctc_blank>""" )
self.assertEqual(len(UpperCAmelCase_ ) , 81 )
def lowerCAmelCase ( self : Optional[int] ) ->int:
"""simple docstring"""
self.assertEqual(self.get_tokenizer().vocab_size , 79 )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = self.get_tokenizers(do_lower_case=UpperCAmelCase_ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
snake_case_ = ["""aaaaa bbbbbb""", """cccccccccdddddddd"""]
snake_case_ = tokenizer.add_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode("""aaaaa bbbbbb low cccccccccdddddddd l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
snake_case_ = {"""eos_token""": """>>>>|||<||<<|<<""", """pad_token""": """<<<<<|||>|>>>>|>"""}
snake_case_ = tokenizer.add_special_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size_a + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode(
""">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->List[str]:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = tokenizer.tokenize("""This is a test""" )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [SPIECE_UNDERLINE, """T""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """a""", SPIECE_UNDERLINE, """t""", """e""", """s""", """t"""] )
# fmt: on
self.assertListEqual(
tokenizer.convert_tokens_to_ids(UpperCAmelCase_ ) , [4, 32, 11, 10, 12, 4, 10, 12, 4, 7, 4, 6, 5, 12, 6] , )
snake_case_ = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """92000""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
snake_case_ = tokenizer.convert_tokens_to_ids(UpperCAmelCase_ )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [4, 30, 4, 20, 7, 12, 4, 25, 8, 13, 9, 4, 10, 9, 4, 3, 23, 4, 7, 9, 14, 4, 6, 11, 10, 12, 4, 10, 12, 4, 19, 7, 15, 12, 73, 26] )
# fmt: on
snake_case_ = tokenizer.convert_ids_to_tokens(UpperCAmelCase_ )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """<unk>""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = [
"""Transformers (formerly known as pytorch-transformers and pytorch-pretrained-bert) provides """
"""general-purpose architectures (BERT, GPT, RoBERTa, XLM, DistilBert, XLNet...) for Natural """
"""Language Understanding (NLU) and Natural Language Generation (NLG) with over thirty-two pretrained """
"""models in one hundred plus languages and deep interoperability between Jax, PyTorch and TensorFlow.""",
"""BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly """
"""conditioning on both left and right context in all layers.""",
"""The quick brown fox jumps over the lazy dog.""",
]
# fmt: off
snake_case_ = {
"""input_ids""": [
[4, 32, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 64, 19, 8, 13, 18, 5, 13, 15, 22, 4, 28, 9, 8, 20, 9, 4, 7, 12, 4, 24, 22, 6, 8, 13, 17, 11, 39, 6, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 7, 9, 14, 4, 24, 22, 6, 8, 13, 17, 11, 39, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 39, 25, 5, 13, 6, 63, 4, 24, 13, 8, 27, 10, 14, 5, 12, 4, 21, 5, 9, 5, 13, 7, 15, 39, 24, 16, 13, 24, 8, 12, 5, 4, 7, 13, 17, 11, 10, 6, 5, 17, 6, 16, 13, 5, 12, 4, 64, 40, 47, 54, 32, 23, 4, 53, 49, 32, 23, 4, 54, 8, 40, 47, 54, 32, 7, 23, 4, 69, 52, 43, 23, 4, 51, 10, 12, 6, 10, 15, 40, 5, 13, 6, 23, 4, 69, 52, 48, 5, 6, 26, 26, 26, 63, 4, 19, 8, 13, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 61, 9, 14, 5, 13, 12, 6, 7, 9, 14, 10, 9, 21, 4, 64, 48, 52, 61, 63, 4, 7, 9, 14, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 53, 5, 9, 5, 13, 7, 6, 10, 8, 9, 4, 64, 48, 52, 53, 63, 4, 20, 10, 6, 11, 4, 8, 27, 5, 13, 4, 6, 11, 10, 13, 6, 22, 39, 6, 20, 8, 4, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 4, 18, 8, 14, 5, 15, 12, 4, 10, 9, 4, 8, 9, 5, 4, 11, 16, 9, 14, 13, 5, 14, 4, 24, 15, 16, 12, 4, 15, 7, 9, 21, 16, 7, 21, 5, 12, 4, 7, 9, 14, 4, 14, 5, 5, 24, 4, 10, 9, 6, 5, 13, 8, 24, 5, 13, 7, 25, 10, 15, 10, 6, 22, 4, 25, 5, 6, 20, 5, 5, 9, 4, 58, 7, 37, 23, 4, 49, 22, 32, 8, 13, 17, 11, 4, 7, 9, 14, 4, 32, 5, 9, 12, 8, 13, 55, 15, 8, 20, 26, 2],
[4, 40, 47, 54, 32, 4, 10, 12, 4, 14, 5, 12, 10, 21, 9, 5, 14, 4, 6, 8, 4, 24, 13, 5, 39, 6, 13, 7, 10, 9, 4, 14, 5, 5, 24, 4, 25, 10, 14, 10, 13, 5, 17, 6, 10, 8, 9, 7, 15, 4, 13, 5, 24, 13, 5, 12, 5, 9, 6, 7, 6, 10, 8, 9, 12, 4, 19, 13, 8, 18, 4, 16, 9, 15, 7, 25, 5, 15, 5, 14, 4, 6, 5, 37, 6, 4, 25, 22, 4, 46, 8, 10, 9, 6, 15, 22, 4, 17, 8, 9, 14, 10, 6, 10, 8, 9, 10, 9, 21, 4, 8, 9, 4, 25, 8, 6, 11, 4, 15, 5, 19, 6, 4, 7, 9, 14, 4, 13, 10, 21, 11, 6, 4, 17, 8, 9, 6, 5, 37, 6, 4, 10, 9, 4, 7, 15, 15, 4, 15, 7, 22, 5, 13, 12, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[4, 32, 11, 5, 4, 45, 16, 10, 17, 28, 4, 25, 13, 8, 20, 9, 4, 19, 8, 37, 4, 46, 16, 18, 24, 12, 4, 8, 27, 5, 13, 4, 6, 11, 5, 4, 15, 7, 57, 22, 4, 14, 8, 21, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
],
"""attention_mask""": [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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],
]
}
# fmt: on
self.tokenizer_integration_test_util(
expected_encoding=UpperCAmelCase_ , model_name="""microsoft/speecht5_asr""" , revision="""c5ef64c71905caeccde0e4462ef3f9077224c524""" , sequences=UpperCAmelCase_ , )
| 347 | 0 |
from urllib.parse import quote
import pytest
from datasets.utils.hub import hf_hub_url
@pytest.mark.parametrize('repo_id' ,['canonical_dataset_name', 'org-name/dataset-name'] )
@pytest.mark.parametrize('path' ,['filename.csv', 'filename with blanks.csv'] )
@pytest.mark.parametrize('revision' ,[None, 'v2'] )
def A ( a_ ,a_ ,a_ ) -> Optional[int]:
__UpperCamelCase : Tuple =hf_hub_url(repo_id=a_ ,path=a_ ,revision=a_ )
assert url == F'https://huggingface.co/datasets/{repo_id}/resolve/{revision or "main"}/{quote(a_ )}'
| 71 |
"""simple docstring"""
import datasets
__SCREAMING_SNAKE_CASE : Tuple = '\\n@InProceedings{conneau2018xnli,\n author = "Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin",\n title = "XNLI: Evaluating Cross-lingual Sentence Representations",\n booktitle = "Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing",\n year = "2018",\n publisher = "Association for Computational Linguistics",\n location = "Brussels, Belgium",\n}\n'
__SCREAMING_SNAKE_CASE : Dict = '\\nXNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n'
__SCREAMING_SNAKE_CASE : List[str] = '\nComputes XNLI score which is just simple accuracy.\nArgs:\n predictions: Predicted labels.\n references: Ground truth labels.\nReturns:\n \'accuracy\': accuracy\nExamples:\n\n >>> predictions = [0, 1]\n >>> references = [0, 1]\n >>> xnli_metric = datasets.load_metric("xnli")\n >>> results = xnli_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {\'accuracy\': 1.0}\n'
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
return (preds == labels).mean()
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION)
class __A (datasets.Metric):
'''simple docstring'''
def lowerCAmelCase ( self : str ) ->Any:
"""simple docstring"""
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
"""predictions""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
"""references""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
} ) , codebase_urls=[] , reference_urls=[] , format="""numpy""" , )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any ) ->int:
"""simple docstring"""
return {"accuracy": simple_accuracy(UpperCAmelCase_ , UpperCAmelCase_ )}
| 347 | 0 |
"""simple docstring"""
from datetime import datetime
import requests
def snake_case_ ( A_ : str ):
'''simple docstring'''
_lowerCamelCase : Union[str, Any] = '''https://downloadgram.net/wp-json/wppress/video-downloader/video?url='''
_lowerCamelCase : Dict = requests.get(base_url + url ).json()[0]['''urls'''][0]['''src''']
return requests.get(A_ ).content
if __name__ == "__main__":
lowerCAmelCase__ = input('''Enter Video/IGTV url: ''').strip()
lowerCAmelCase__ = F"""{datetime.now():%Y-%m-%d_%H:%M:%S}.mp4"""
with open(file_name, '''wb''') as fp:
fp.write(download_video(url))
print(F"""Done. Video saved to disk as {file_name}.""")
| 72 |
"""simple docstring"""
from ..utils import DummyObject, requires_backends
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Tuple ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : int ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Dict ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Union[str, Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[Any] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : str , **UpperCAmelCase_ : int ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : Union[str, Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[Any] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : List[str] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Optional[Any] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Optional[Any] ) ->int:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Dict ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Union[str, Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[int] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Optional[int] ) ->Any:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
| 347 | 0 |
import argparse
import math
import os
import torch
from neural_compressor.utils.pytorch import load
from PIL import Image
from transformers import CLIPTextModel, CLIPTokenizer
from diffusers import AutoencoderKL, StableDiffusionPipeline, UNetaDConditionModel
def SCREAMING_SNAKE_CASE__ ( ) -> Optional[Any]:
__lowerCamelCase : Tuple = argparse.ArgumentParser()
parser.add_argument(
'-m' , '--pretrained_model_name_or_path' , type=lowerCamelCase__ , default=lowerCamelCase__ , required=lowerCamelCase__ , help='Path to pretrained model or model identifier from huggingface.co/models.' , )
parser.add_argument(
'-c' , '--caption' , type=lowerCamelCase__ , default='robotic cat with wings' , help='Text used to generate images.' , )
parser.add_argument(
'-n' , '--images_num' , type=lowerCamelCase__ , default=4 , help='How much images to generate.' , )
parser.add_argument(
'-s' , '--seed' , type=lowerCamelCase__ , default=4_2 , help='Seed for random process.' , )
parser.add_argument(
'-ci' , '--cuda_id' , type=lowerCamelCase__ , default=0 , help='cuda_id.' , )
__lowerCamelCase : Any = parser.parse_args()
return args
def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ ) -> Any:
if not len(lowerCamelCase__ ) == rows * cols:
raise ValueError('The specified number of rows and columns are not correct.' )
__lowerCamelCase , __lowerCamelCase : Optional[int] = imgs[0].size
__lowerCamelCase : List[Any] = Image.new('RGB' , size=(cols * w, rows * h) )
__lowerCamelCase , __lowerCamelCase : List[str] = grid.size
for i, img in enumerate(lowerCamelCase__ ):
grid.paste(lowerCamelCase__ , box=(i % cols * w, i // cols * h) )
return grid
def SCREAMING_SNAKE_CASE__ ( lowerCamelCase__ , lowerCamelCase__="robotic cat with wings" , lowerCamelCase__=7.5 , lowerCamelCase__=5_0 , lowerCamelCase__=1 , lowerCamelCase__=4_2 , ) -> int:
__lowerCamelCase : Any = torch.Generator(pipeline.device ).manual_seed(lowerCamelCase__ )
__lowerCamelCase : int = pipeline(
lowerCamelCase__ , guidance_scale=lowerCamelCase__ , num_inference_steps=lowerCamelCase__ , generator=lowerCamelCase__ , num_images_per_prompt=lowerCamelCase__ , ).images
__lowerCamelCase : Union[str, Any] = int(math.sqrt(lowerCamelCase__ ) )
__lowerCamelCase : str = image_grid(lowerCamelCase__ , rows=_rows , cols=num_images_per_prompt // _rows )
return grid, images
a =parse_args()
# Load models and create wrapper for stable diffusion
a =CLIPTokenizer.from_pretrained(args.pretrained_model_name_or_path, subfolder="""tokenizer""")
a =CLIPTextModel.from_pretrained(args.pretrained_model_name_or_path, subfolder="""text_encoder""")
a =AutoencoderKL.from_pretrained(args.pretrained_model_name_or_path, subfolder="""vae""")
a =UNetaDConditionModel.from_pretrained(args.pretrained_model_name_or_path, subfolder="""unet""")
a =StableDiffusionPipeline.from_pretrained(
args.pretrained_model_name_or_path, text_encoder=text_encoder, vae=vae, unet=unet, tokenizer=tokenizer
)
a =lambda images, clip_input: (images, False)
if os.path.exists(os.path.join(args.pretrained_model_name_or_path, """best_model.pt""")):
a =load(args.pretrained_model_name_or_path, model=unet)
unet.eval()
setattr(pipeline, """unet""", unet)
else:
a =unet.to(torch.device("""cuda""", args.cuda_id))
a =pipeline.to(unet.device)
a , a =generate_images(pipeline, prompt=args.caption, num_images_per_prompt=args.images_num, seed=args.seed)
grid.save(os.path.join(args.pretrained_model_name_or_path, """{}.png""".format("""_""".join(args.caption.split()))))
a =os.path.join(args.pretrained_model_name_or_path, """_""".join(args.caption.split()))
os.makedirs(dirname, exist_ok=True)
for idx, image in enumerate(images):
image.save(os.path.join(dirname, """{}.png""".format(idx + 1)))
| 73 |
"""simple docstring"""
import warnings
from ...utils import logging
from .image_processing_mobilevit import MobileViTImageProcessor
__SCREAMING_SNAKE_CASE : int = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
def __init__( self : str , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->None:
"""simple docstring"""
warnings.warn(
"""The class MobileViTFeatureExtractor is deprecated and will be removed in version 5 of Transformers."""
""" Please use MobileViTImageProcessor instead.""" , UpperCAmelCase_ , )
super().__init__(*UpperCAmelCase_ , **UpperCAmelCase_ )
| 347 | 0 |
"""simple docstring"""
import gc
import random
import unittest
import numpy as np
import torch
from transformers import CLIPTextConfig, CLIPTextModel, CLIPTokenizer
import diffusers
from diffusers import (
AutoencoderKL,
EulerDiscreteScheduler,
StableDiffusionLatentUpscalePipeline,
StableDiffusionPipeline,
UNetaDConditionModel,
)
from diffusers.schedulers import KarrasDiffusionSchedulers
from diffusers.utils import floats_tensor, load_image, load_numpy, slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..pipeline_params import TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS, TEXT_GUIDED_IMAGE_VARIATION_PARAMS
from ..test_pipelines_common import PipelineKarrasSchedulerTesterMixin, PipelineLatentTesterMixin, PipelineTesterMixin
enable_full_determinism()
def _snake_case ( snake_case__ : Union[str, Any] ):
A = [tensor.shape for tensor in tensor_list]
return all(shape == shapes[0] for shape in shapes[1:] )
class lowerCAmelCase_ ( _lowercase , _lowercase , _lowercase , unittest.TestCase ):
'''simple docstring'''
_lowerCamelCase: Any = StableDiffusionLatentUpscalePipeline
_lowerCamelCase: Dict = TEXT_GUIDED_IMAGE_VARIATION_PARAMS - {
'''height''',
'''width''',
'''cross_attention_kwargs''',
'''negative_prompt_embeds''',
'''prompt_embeds''',
}
_lowerCamelCase: List[str] = PipelineTesterMixin.required_optional_params - {'''num_images_per_prompt'''}
_lowerCamelCase: int = TEXT_GUIDED_IMAGE_VARIATION_BATCH_PARAMS
_lowerCamelCase: Union[str, Any] = frozenset(
[] ) # TO-DO: update image_params once pipeline is refactored with VaeImageProcessor.preprocess
_lowerCamelCase: Optional[int] = frozenset([] )
_lowerCamelCase: Tuple = True
@property
def _SCREAMING_SNAKE_CASE ( self : str ) -> str:
A = 1
A = 4
A = (16, 16)
A = floats_tensor((batch_size, num_channels) + sizes ,rng=random.Random(0 ) ).to(A_ )
return image
def _SCREAMING_SNAKE_CASE ( self : int ) -> str:
torch.manual_seed(0 )
A = UNetaDConditionModel(
act_fn='gelu' ,attention_head_dim=8 ,norm_num_groups=A_ ,block_out_channels=[32, 32, 64, 64] ,time_cond_proj_dim=160 ,conv_in_kernel=1 ,conv_out_kernel=1 ,cross_attention_dim=32 ,down_block_types=(
'KDownBlock2D',
'KCrossAttnDownBlock2D',
'KCrossAttnDownBlock2D',
'KCrossAttnDownBlock2D',
) ,in_channels=8 ,mid_block_type=A_ ,only_cross_attention=A_ ,out_channels=5 ,resnet_time_scale_shift='scale_shift' ,time_embedding_type='fourier' ,timestep_post_act='gelu' ,up_block_types=('KCrossAttnUpBlock2D', 'KCrossAttnUpBlock2D', 'KCrossAttnUpBlock2D', 'KUpBlock2D') ,)
A = AutoencoderKL(
block_out_channels=[32, 32, 64, 64] ,in_channels=3 ,out_channels=3 ,down_block_types=[
'DownEncoderBlock2D',
'DownEncoderBlock2D',
'DownEncoderBlock2D',
'DownEncoderBlock2D',
] ,up_block_types=['UpDecoderBlock2D', 'UpDecoderBlock2D', 'UpDecoderBlock2D', 'UpDecoderBlock2D'] ,latent_channels=4 ,)
A = EulerDiscreteScheduler(prediction_type='sample' )
A = CLIPTextConfig(
bos_token_id=0 ,eos_token_id=2 ,hidden_size=32 ,intermediate_size=37 ,layer_norm_eps=1e-05 ,num_attention_heads=4 ,num_hidden_layers=5 ,pad_token_id=1 ,vocab_size=1000 ,hidden_act='quick_gelu' ,projection_dim=512 ,)
A = CLIPTextModel(A_ )
A = CLIPTokenizer.from_pretrained('hf-internal-testing/tiny-random-clip' )
A = {
'unet': model.eval(),
'vae': vae.eval(),
'scheduler': scheduler,
'text_encoder': text_encoder,
'tokenizer': tokenizer,
}
return components
def _SCREAMING_SNAKE_CASE ( self : Any ,A_ : Optional[int] ,A_ : Union[str, Any]=0 ) -> List[Any]:
if str(A_ ).startswith('mps' ):
A = torch.manual_seed(A_ )
else:
A = torch.Generator(device=A_ ).manual_seed(A_ )
A = {
'prompt': 'A painting of a squirrel eating a burger',
'image': self.dummy_image.cpu(),
'generator': generator,
'num_inference_steps': 2,
'output_type': 'numpy',
}
return inputs
def _SCREAMING_SNAKE_CASE ( self : Dict ) -> str:
A = 'cpu'
A = self.get_dummy_components()
A = self.pipeline_class(**A_ )
pipe.to(A_ )
pipe.set_progress_bar_config(disable=A_ )
A = self.get_dummy_inputs(A_ )
A = pipe(**A_ ).images
A = image[0, -3:, -3:, -1]
self.assertEqual(image.shape ,(1, 256, 256, 3) )
A = np.array(
[0.47_22_24_12, 0.41_92_16_33, 0.44_71_74_34, 0.46_87_41_92, 0.42_58_82_58, 0.46_15_07_26, 0.4_67_75_34, 0.45_58_38_32, 0.48_57_90_55] )
A = np.abs(image_slice.flatten() - expected_slice ).max()
self.assertLessEqual(A_ ,1e-3 )
def _SCREAMING_SNAKE_CASE ( self : Optional[int] ) -> Any:
super().test_attention_slicing_forward_pass(expected_max_diff=7e-3 )
def _SCREAMING_SNAKE_CASE ( self : Union[str, Any] ) -> Optional[Any]:
super().test_cpu_offload_forward_pass(expected_max_diff=3e-3 )
def _SCREAMING_SNAKE_CASE ( self : Optional[int] ) -> Dict:
super().test_dict_tuple_outputs_equivalent(expected_max_difference=3e-3 )
def _SCREAMING_SNAKE_CASE ( self : int ) -> Optional[int]:
super().test_inference_batch_single_identical(expected_max_diff=7e-3 )
def _SCREAMING_SNAKE_CASE ( self : str ) -> Dict:
super().test_pt_np_pil_outputs_equivalent(expected_max_diff=3e-3 )
def _SCREAMING_SNAKE_CASE ( self : Optional[int] ) -> Any:
super().test_save_load_local(expected_max_difference=3e-3 )
def _SCREAMING_SNAKE_CASE ( self : str ) -> Dict:
super().test_save_load_optional_components(expected_max_difference=3e-3 )
def _SCREAMING_SNAKE_CASE ( self : Dict ) -> Any:
A = [
'DDIMScheduler',
'DDPMScheduler',
'PNDMScheduler',
'HeunDiscreteScheduler',
'EulerAncestralDiscreteScheduler',
'KDPM2DiscreteScheduler',
'KDPM2AncestralDiscreteScheduler',
'DPMSolverSDEScheduler',
]
A = self.get_dummy_components()
A = self.pipeline_class(**A_ )
# make sure that PNDM does not need warm-up
pipe.scheduler.register_to_config(skip_prk_steps=A_ )
pipe.to(A_ )
pipe.set_progress_bar_config(disable=A_ )
A = self.get_dummy_inputs(A_ )
A = 2
A = []
for scheduler_enum in KarrasDiffusionSchedulers:
if scheduler_enum.name in skip_schedulers:
# no sigma schedulers are not supported
# no schedulers
continue
A = getattr(A_ ,scheduler_enum.name )
A = scheduler_cls.from_config(pipe.scheduler.config )
A = pipe(**A_ )[0]
outputs.append(A_ )
assert check_same_shape(A_ )
@require_torch_gpu
@slow
class lowerCAmelCase_ ( unittest.TestCase ):
'''simple docstring'''
def _SCREAMING_SNAKE_CASE ( self : Union[str, Any] ) -> int:
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def _SCREAMING_SNAKE_CASE ( self : Optional[Any] ) -> int:
A = torch.manual_seed(33 )
A = StableDiffusionPipeline.from_pretrained('CompVis/stable-diffusion-v1-4' ,torch_dtype=torch.floataa )
pipe.to('cuda' )
A = StableDiffusionLatentUpscalePipeline.from_pretrained(
'stabilityai/sd-x2-latent-upscaler' ,torch_dtype=torch.floataa )
upscaler.to('cuda' )
A = 'a photo of an astronaut high resolution, unreal engine, ultra realistic'
A = pipe(A_ ,generator=A_ ,output_type='latent' ).images
A = upscaler(
prompt=A_ ,image=A_ ,num_inference_steps=20 ,guidance_scale=0 ,generator=A_ ,output_type='np' ,).images[0]
A = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/latent-upscaler/astronaut_1024.npy' )
assert np.abs((expected_image - image).mean() ) < 5e-2
def _SCREAMING_SNAKE_CASE ( self : List[str] ) -> Optional[Any]:
A = torch.manual_seed(33 )
A = StableDiffusionLatentUpscalePipeline.from_pretrained(
'stabilityai/sd-x2-latent-upscaler' ,torch_dtype=torch.floataa )
upscaler.to('cuda' )
A = 'the temple of fire by Ross Tran and Gerardo Dottori, oil on canvas'
A = load_image(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/latent-upscaler/fire_temple_512.png' )
A = upscaler(
prompt=A_ ,image=A_ ,num_inference_steps=20 ,guidance_scale=0 ,generator=A_ ,output_type='np' ,).images[0]
A = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/latent-upscaler/fire_temple_1024.npy' )
assert np.abs((expected_image - image).max() ) < 5e-2 | 74 |
"""simple docstring"""
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Any:
snake_case_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
("""cls_token""", """vit.embeddings.cls_token"""),
("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight"""),
("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias"""),
("""pos_embed""", """vit.embeddings.position_embeddings"""),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("""norm.weight""", """layernorm.weight"""),
("""norm.bias""", """layernorm.bias"""),
("""pre_logits.fc.weight""", """pooler.dense.weight"""),
("""pre_logits.fc.bias""", """pooler.dense.bias"""),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
snake_case_ = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("""norm.weight""", """vit.layernorm.weight"""),
("""norm.bias""", """vit.layernorm.bias"""),
("""head.weight""", """classifier.weight"""),
("""head.bias""", """classifier.bias"""),
] )
return rename_keys
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
snake_case_ = """"""
else:
snake_case_ = """vit."""
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case_ = in_proj_weight[
: config.hidden_size, :
]
snake_case_ = in_proj_bias[: config.hidden_size]
snake_case_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case_ = in_proj_weight[
-config.hidden_size :, :
]
snake_case_ = in_proj_bias[-config.hidden_size :]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = ["""head.weight""", """head.bias"""]
for k in ignore_keys:
state_dict.pop(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> str:
snake_case_ = dct.pop(_SCREAMING_SNAKE_CASE )
snake_case_ = val
def _a ( ) -> Any:
snake_case_ = """http://images.cocodataset.org/val2017/000000039769.jpg"""
snake_case_ = Image.open(requests.get(_SCREAMING_SNAKE_CASE , stream=_SCREAMING_SNAKE_CASE ).raw )
return im
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = ViTConfig()
snake_case_ = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
snake_case_ = True
snake_case_ = int(vit_name[-12:-10] )
snake_case_ = int(vit_name[-9:-6] )
else:
snake_case_ = 1_000
snake_case_ = """huggingface/label-files"""
snake_case_ = """imagenet-1k-id2label.json"""
snake_case_ = json.load(open(hf_hub_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , repo_type="""dataset""" ) , """r""" ) )
snake_case_ = {int(_SCREAMING_SNAKE_CASE ): v for k, v in idalabel.items()}
snake_case_ = idalabel
snake_case_ = {v: k for k, v in idalabel.items()}
snake_case_ = int(vit_name[-6:-4] )
snake_case_ = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith("""tiny""" ):
snake_case_ = 192
snake_case_ = 768
snake_case_ = 12
snake_case_ = 3
elif vit_name[9:].startswith("""small""" ):
snake_case_ = 384
snake_case_ = 1_536
snake_case_ = 12
snake_case_ = 6
else:
pass
else:
if vit_name[4:].startswith("""small""" ):
snake_case_ = 768
snake_case_ = 2_304
snake_case_ = 8
snake_case_ = 8
elif vit_name[4:].startswith("""base""" ):
pass
elif vit_name[4:].startswith("""large""" ):
snake_case_ = 1_024
snake_case_ = 4_096
snake_case_ = 24
snake_case_ = 16
elif vit_name[4:].startswith("""huge""" ):
snake_case_ = 1_280
snake_case_ = 5_120
snake_case_ = 32
snake_case_ = 16
# load original model from timm
snake_case_ = timm.create_model(_SCREAMING_SNAKE_CASE , pretrained=_SCREAMING_SNAKE_CASE )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case_ = timm_model.state_dict()
if base_model:
remove_classification_head_(_SCREAMING_SNAKE_CASE )
snake_case_ = create_rename_keys(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
for src, dest in rename_keys:
rename_key(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
read_in_q_k_v(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# load HuggingFace model
if vit_name[-5:] == "in21k":
snake_case_ = ViTModel(_SCREAMING_SNAKE_CASE ).eval()
else:
snake_case_ = ViTForImageClassification(_SCREAMING_SNAKE_CASE ).eval()
model.load_state_dict(_SCREAMING_SNAKE_CASE )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
snake_case_ = DeiTImageProcessor(size=config.image_size )
else:
snake_case_ = ViTImageProcessor(size=config.image_size )
snake_case_ = image_processor(images=prepare_img() , return_tensors="""pt""" )
snake_case_ = encoding["""pixel_values"""]
snake_case_ = model(_SCREAMING_SNAKE_CASE )
if base_model:
snake_case_ = timm_model.forward_features(_SCREAMING_SNAKE_CASE )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.pooler_output , atol=1E-3 )
else:
snake_case_ = timm_model(_SCREAMING_SNAKE_CASE )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.logits , atol=1E-3 )
Path(_SCREAMING_SNAKE_CASE ).mkdir(exist_ok=_SCREAMING_SNAKE_CASE )
print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_SCREAMING_SNAKE_CASE )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--vit_name',
default='vit_base_patch16_224',
type=str,
help='Name of the ViT timm model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
__SCREAMING_SNAKE_CASE : int = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 347 | 0 |
'''simple docstring'''
import unittest
from parameterized import parameterized
from transformers import LlamaConfig, is_torch_available, set_seed
from transformers.testing_utils import require_torch, slow, torch_device
from ...generation.test_utils import GenerationTesterMixin
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, ids_tensor, random_attention_mask
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from transformers import LlamaForCausalLM, LlamaForSequenceClassification, LlamaModel, LlamaTokenizer
class __UpperCamelCase :
def __init__( self, lowerCAmelCase, lowerCAmelCase=13, lowerCAmelCase=7, lowerCAmelCase=True, lowerCAmelCase=True, lowerCAmelCase=False, lowerCAmelCase=True, lowerCAmelCase=99, lowerCAmelCase=32, lowerCAmelCase=5, lowerCAmelCase=4, lowerCAmelCase=37, lowerCAmelCase="gelu", lowerCAmelCase=0.1, lowerCAmelCase=0.1, lowerCAmelCase=512, lowerCAmelCase=16, lowerCAmelCase=2, lowerCAmelCase=0.0_2, lowerCAmelCase=3, lowerCAmelCase=4, lowerCAmelCase=None, ):
"""simple docstring"""
lowerCamelCase_ =parent
lowerCamelCase_ =batch_size
lowerCamelCase_ =seq_length
lowerCamelCase_ =is_training
lowerCamelCase_ =use_input_mask
lowerCamelCase_ =use_token_type_ids
lowerCamelCase_ =use_labels
lowerCamelCase_ =vocab_size
lowerCamelCase_ =hidden_size
lowerCamelCase_ =num_hidden_layers
lowerCamelCase_ =num_attention_heads
lowerCamelCase_ =intermediate_size
lowerCamelCase_ =hidden_act
lowerCamelCase_ =hidden_dropout_prob
lowerCamelCase_ =attention_probs_dropout_prob
lowerCamelCase_ =max_position_embeddings
lowerCamelCase_ =type_vocab_size
lowerCamelCase_ =type_sequence_label_size
lowerCamelCase_ =initializer_range
lowerCamelCase_ =num_labels
lowerCamelCase_ =num_choices
lowerCamelCase_ =scope
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =ids_tensor([self.batch_size, self.seq_length], self.vocab_size )
lowerCamelCase_ =None
if self.use_input_mask:
lowerCamelCase_ =random_attention_mask([self.batch_size, self.seq_length] )
lowerCamelCase_ =None
if self.use_token_type_ids:
lowerCamelCase_ =ids_tensor([self.batch_size, self.seq_length], self.type_vocab_size )
lowerCamelCase_ =None
lowerCamelCase_ =None
lowerCamelCase_ =None
if self.use_labels:
lowerCamelCase_ =ids_tensor([self.batch_size], self.type_sequence_label_size )
lowerCamelCase_ =ids_tensor([self.batch_size, self.seq_length], self.num_labels )
lowerCamelCase_ =ids_tensor([self.batch_size], self.num_choices )
lowerCamelCase_ =self.get_config()
return config, input_ids, token_type_ids, input_mask, sequence_labels, token_labels, choice_labels
def lowercase__ ( self ):
"""simple docstring"""
return LlamaConfig(
vocab_size=self.vocab_size, hidden_size=self.hidden_size, num_hidden_layers=self.num_hidden_layers, num_attention_heads=self.num_attention_heads, intermediate_size=self.intermediate_size, hidden_act=self.hidden_act, hidden_dropout_prob=self.hidden_dropout_prob, attention_probs_dropout_prob=self.attention_probs_dropout_prob, max_position_embeddings=self.max_position_embeddings, type_vocab_size=self.type_vocab_size, is_decoder=lowerCAmelCase, initializer_range=self.initializer_range, )
def lowercase__ ( self, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase ):
"""simple docstring"""
lowerCamelCase_ =LlamaModel(config=lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase )
lowerCamelCase_ =model(lowerCAmelCase )
self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.seq_length, self.hidden_size) )
def lowercase__ ( self, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, ):
"""simple docstring"""
lowerCamelCase_ =True
lowerCamelCase_ =LlamaModel(lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(
lowerCAmelCase, attention_mask=lowerCAmelCase, encoder_hidden_states=lowerCAmelCase, encoder_attention_mask=lowerCAmelCase, )
lowerCamelCase_ =model(
lowerCAmelCase, attention_mask=lowerCAmelCase, encoder_hidden_states=lowerCAmelCase, )
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase )
self.parent.assertEqual(result.last_hidden_state.shape, (self.batch_size, self.seq_length, self.hidden_size) )
def lowercase__ ( self, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, ):
"""simple docstring"""
lowerCamelCase_ =LlamaForCausalLM(config=lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase, labels=lowerCAmelCase )
self.parent.assertEqual(result.logits.shape, (self.batch_size, self.seq_length, self.vocab_size) )
def lowercase__ ( self, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, lowerCAmelCase, ):
"""simple docstring"""
lowerCamelCase_ =True
lowerCamelCase_ =True
lowerCamelCase_ =LlamaForCausalLM(config=lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
# first forward pass
lowerCamelCase_ =model(
lowerCAmelCase, attention_mask=lowerCAmelCase, encoder_hidden_states=lowerCAmelCase, encoder_attention_mask=lowerCAmelCase, use_cache=lowerCAmelCase, )
lowerCamelCase_ =outputs.past_key_values
# create hypothetical multiple next token and extent to next_input_ids
lowerCamelCase_ =ids_tensor((self.batch_size, 3), config.vocab_size )
lowerCamelCase_ =ids_tensor((self.batch_size, 3), vocab_size=2 )
# append to next input_ids and
lowerCamelCase_ =torch.cat([input_ids, next_tokens], dim=-1 )
lowerCamelCase_ =torch.cat([input_mask, next_mask], dim=-1 )
lowerCamelCase_ =model(
lowerCAmelCase, attention_mask=lowerCAmelCase, encoder_hidden_states=lowerCAmelCase, encoder_attention_mask=lowerCAmelCase, output_hidden_states=lowerCAmelCase, )['''hidden_states'''][0]
lowerCamelCase_ =model(
lowerCAmelCase, attention_mask=lowerCAmelCase, encoder_hidden_states=lowerCAmelCase, encoder_attention_mask=lowerCAmelCase, past_key_values=lowerCAmelCase, output_hidden_states=lowerCAmelCase, )['''hidden_states'''][0]
# select random slice
lowerCamelCase_ =ids_tensor((1,), output_from_past.shape[-1] ).item()
lowerCamelCase_ =output_from_no_past[:, -3:, random_slice_idx].detach()
lowerCamelCase_ =output_from_past[:, :, random_slice_idx].detach()
self.parent.assertTrue(output_from_past_slice.shape[1] == next_tokens.shape[1] )
# test that outputs are equal for slice
self.parent.assertTrue(torch.allclose(lowerCAmelCase, lowerCAmelCase, atol=1e-3 ) )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.prepare_config_and_inputs()
(
(
lowerCamelCase_
), (
lowerCamelCase_
), (
lowerCamelCase_
), (
lowerCamelCase_
), (
lowerCamelCase_
), (
lowerCamelCase_
), (
lowerCamelCase_
),
) =config_and_inputs
lowerCamelCase_ ={'''input_ids''': input_ids, '''attention_mask''': input_mask}
return config, inputs_dict
@require_torch
class __UpperCamelCase ( lowerCamelCase__ , lowerCamelCase__ , lowerCamelCase__ , unittest.TestCase ):
lowercase : str =(LlamaModel, LlamaForCausalLM, LlamaForSequenceClassification) if is_torch_available() else ()
lowercase : Optional[Any] =(LlamaForCausalLM,) if is_torch_available() else ()
lowercase : int =(
{
'feature-extraction': LlamaModel,
'text-classification': LlamaForSequenceClassification,
'text-generation': LlamaForCausalLM,
'zero-shot': LlamaForSequenceClassification,
}
if is_torch_available()
else {}
)
lowercase : str =False
lowercase : List[Any] =False
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =LlamaModelTester(self )
lowerCamelCase_ =ConfigTester(self, config_class=lowerCAmelCase, hidden_size=37 )
def lowercase__ ( self ):
"""simple docstring"""
self.config_tester.run_common_tests()
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*lowerCAmelCase )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =self.model_tester.prepare_config_and_inputs()
for type in ["absolute", "relative_key", "relative_key_query"]:
lowerCamelCase_ =type
self.model_tester.create_and_check_model(*lowerCAmelCase )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_, lowerCamelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
lowerCamelCase_ =3
lowerCamelCase_ =input_dict['''input_ids''']
lowerCamelCase_ =input_ids.ne(1 ).to(lowerCAmelCase )
lowerCamelCase_ =ids_tensor([self.model_tester.batch_size], self.model_tester.type_sequence_label_size )
lowerCamelCase_ =LlamaForSequenceClassification(lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase, labels=lowerCAmelCase )
self.assertEqual(result.logits.shape, (self.model_tester.batch_size, self.model_tester.num_labels) )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_, lowerCamelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
lowerCamelCase_ =3
lowerCamelCase_ ='''single_label_classification'''
lowerCamelCase_ =input_dict['''input_ids''']
lowerCamelCase_ =input_ids.ne(1 ).to(lowerCAmelCase )
lowerCamelCase_ =ids_tensor([self.model_tester.batch_size], self.model_tester.type_sequence_label_size )
lowerCamelCase_ =LlamaForSequenceClassification(lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase, labels=lowerCAmelCase )
self.assertEqual(result.logits.shape, (self.model_tester.batch_size, self.model_tester.num_labels) )
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_, lowerCamelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
lowerCamelCase_ =3
lowerCamelCase_ ='''multi_label_classification'''
lowerCamelCase_ =input_dict['''input_ids''']
lowerCamelCase_ =input_ids.ne(1 ).to(lowerCAmelCase )
lowerCamelCase_ =ids_tensor(
[self.model_tester.batch_size, config.num_labels], self.model_tester.type_sequence_label_size ).to(torch.float )
lowerCamelCase_ =LlamaForSequenceClassification(lowerCAmelCase )
model.to(lowerCAmelCase )
model.eval()
lowerCamelCase_ =model(lowerCAmelCase, attention_mask=lowerCAmelCase, labels=lowerCAmelCase )
self.assertEqual(result.logits.shape, (self.model_tester.batch_size, self.model_tester.num_labels) )
@unittest.skip('''LLaMA buffers include complex numbers, which breaks this test''' )
def lowercase__ ( self ):
"""simple docstring"""
pass
@parameterized.expand([('''linear''',), ('''dynamic''',)] )
def lowercase__ ( self, lowerCAmelCase ):
"""simple docstring"""
lowerCamelCase_, lowerCamelCase_ =self.model_tester.prepare_config_and_inputs_for_common()
lowerCamelCase_ =ids_tensor([1, 10], config.vocab_size )
lowerCamelCase_ =ids_tensor([1, int(config.max_position_embeddings * 1.5 )], config.vocab_size )
set_seed(42 ) # Fixed seed at init time so the two models get the same random weights
lowerCamelCase_ =LlamaModel(lowerCAmelCase )
original_model.to(lowerCAmelCase )
original_model.eval()
lowerCamelCase_ =original_model(lowerCAmelCase ).last_hidden_state
lowerCamelCase_ =original_model(lowerCAmelCase ).last_hidden_state
set_seed(42 ) # Fixed seed at init time so the two models get the same random weights
lowerCamelCase_ ={'''type''': scaling_type, '''factor''': 1_0.0}
lowerCamelCase_ =LlamaModel(lowerCAmelCase )
scaled_model.to(lowerCAmelCase )
scaled_model.eval()
lowerCamelCase_ =scaled_model(lowerCAmelCase ).last_hidden_state
lowerCamelCase_ =scaled_model(lowerCAmelCase ).last_hidden_state
# Dynamic scaling does not change the RoPE embeddings until it receives an input longer than the original
# maximum sequence length, so the outputs for the short input should match.
if scaling_type == "dynamic":
self.assertTrue(torch.allclose(lowerCAmelCase, lowerCAmelCase, atol=1e-5 ) )
else:
self.assertFalse(torch.allclose(lowerCAmelCase, lowerCAmelCase, atol=1e-5 ) )
# The output should be different for long inputs
self.assertFalse(torch.allclose(lowerCAmelCase, lowerCAmelCase, atol=1e-5 ) )
@require_torch
class __UpperCamelCase ( unittest.TestCase ):
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''' )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =[1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
lowerCamelCase_ =LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-7b-hf''', device_map='''auto''' )
lowerCamelCase_ =model(torch.tensor([input_ids] ) )
# Expected mean on dim = -1
lowerCamelCase_ =torch.tensor([[-6.6_5_5_0, -4.1_2_2_7, -4.9_8_5_9, -3.2_4_0_6, 0.8_2_6_2, -3.0_0_3_3, 1.2_9_6_4, -3.3_6_9_9]] )
torch.testing.assert_close(out.mean(-1 ), lowerCAmelCase, atol=1e-2, rtol=1e-2 )
# slicing logits[0, 0, 0:30]
# fmt: off
lowerCamelCase_ =torch.tensor([-1_2.8_2_8_1, -7.4_4_5_3, -0.4_6_3_9, -8.0_6_2_5, -7.2_5_0_0, -8.0_0_0_0, -6.4_8_8_3, -7.7_6_9_5, -7.8_4_3_8, -7.0_3_1_2, -6.2_1_8_8, -7.1_3_2_8, -1.8_4_9_6, 1.9_9_6_1, -8.6_2_5_0, -6.7_2_2_7, -1_2.8_2_8_1, -6.9_4_9_2, -7.0_7_4_2, -7.7_8_5_2, -7.5_8_2_0, -7.9_0_6_2, -6.9_3_7_5, -7.9_8_0_5, -8.3_4_3_8, -8.1_5_6_2, -8.0_4_6_9, -7.6_2_5_0, -7.7_4_2_2, -7.3_3_9_8,] )
# fmt: on
torch.testing.assert_close(out[0, 0, :30], lowerCAmelCase, atol=1e-5, rtol=1e-5 )
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''' )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =[1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
lowerCamelCase_ =LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-13b-hf''', device_map='''auto''' )
lowerCamelCase_ =model(torch.tensor(lowerCAmelCase ) )
# Expected mean on dim = -1
lowerCamelCase_ =torch.tensor([[-2.0_6_2_2, -1.2_7_9_4, -1.1_6_3_8, -0.9_7_8_8, -1.4_6_0_3, -1.0_2_3_8, -1.7_8_9_3, -1.4_4_1_1]] )
torch.testing.assert_close(out.mean(-1 ), lowerCAmelCase, atol=1e-2, rtol=1e-2 )
# slicing logits[0, 0, 0:30]
# fmt: off
lowerCamelCase_ =torch.tensor([-8.1_4_0_6, -8.0_5_4_7, 2.7_4_6_1, -1.2_3_4_4, -0.1_4_4_8, -1.8_2_6_2, -1.0_0_2_0, -1.8_1_5_4, -1.6_8_9_5, -1.8_5_1_6, -2.3_5_7_4, -0.9_2_7_7, 3.7_5_9_8, 6.5_7_4_2, -1.2_9_9_8, -0.1_1_7_7, -8.1_4_0_6, -2.9_6_8_8, -2.9_1_9_9, -3.1_6_9_9, -3.5_2_5_4, -2.3_5_5_5, -2.7_9_8_8, -3.4_1_4_1, -2.8_2_6_2, -4.5_1_9_5, -3.3_3_7_9, -3.3_1_6_4, -2.7_8_3_2, -3.0_2_7_3] )
# fmt: on
torch.testing.assert_close(out[0, 0, :30], lowerCAmelCase, atol=1e-5, rtol=1e-5 )
@unittest.skip('''Logits are not exactly the same, once we fix the instabalities somehow, will update!''' )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =[1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
lowerCamelCase_ =LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-13b-chat-hf''', device_map='''auto''' )
lowerCamelCase_ =model(torch.tensor(lowerCAmelCase ) )
# Expected mean on dim = -1
lowerCamelCase_ =torch.tensor([[-0.8_5_6_2, -1.8_5_2_0, -0.7_5_5_1, -0.4_1_6_2, -1.5_1_6_1, -1.2_0_3_8, -2.4_8_2_3, -2.3_2_5_4]] )
torch.testing.assert_close(out.mean(-1 ), lowerCAmelCase, atol=1e-2, rtol=1e-2 )
# slicing logits[0, 0, 0:30]
# fmt: off
lowerCamelCase_ =torch.tensor([-2.2_2_2_7, 4.8_8_2_8, 0.9_0_2_3, -0.4_5_7_8, -0.7_8_7_1, -0.1_0_3_3, -0.6_2_2_1, -0.5_7_8_6, -0.7_8_0_3, -1.0_6_7_4, -1.2_9_2_0, -0.1_5_7_0, 0.8_0_0_8, 2.0_7_2_3, -0.9_4_9_7, 0.2_7_7_1, -2.2_2_2_7, -0.7_6_1_2, -1.4_3_4_6, -1.2_0_6_1, -1.6_4_2_6, -0.3_0_0_0, -0.7_1_3_9, -1.1_9_3_4, -1.8_6_9_1, -1.6_9_7_3, -1.5_9_4_7, -1.2_7_0_5, -0.3_5_2_3, -0.5_5_1_3] )
# fmt: on
torch.testing.assert_close(out.mean(-1 ), lowerCAmelCase, atol=1e-2, rtol=1e-2 )
@unittest.skip(
'''Logits are not exactly the same, once we fix the instabalities somehow, will update! Also it is gonna be a `too_slow` test''' )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ =[1, 306, 4_658, 278, 6_593, 310, 2_834, 338]
lowerCamelCase_ =LlamaForCausalLM.from_pretrained('''meta-llama/Llama-2-70b-hf''', device_map='''auto''' )
lowerCamelCase_ =model(torch.tensor(lowerCAmelCase ) )
lowerCamelCase_ =torch.tensor(
[[-4.2_3_2_7, -3.3_3_6_0, -4.6_6_6_5, -4.7_6_3_1, -1.8_1_8_0, -3.4_1_7_0, -1.4_2_1_1, -3.1_8_1_0]], dtype=torch.floataa )
torch.testing.assert_close(out.mean(-1 ), lowerCAmelCase, atol=1e-2, rtol=1e-2 )
# fmt: off
lowerCamelCase_ =torch.tensor([-9.4_9_2_2, -3.9_5_5_1, 1.7_9_9_8, -5.6_7_5_8, -5.1_0_5_5, -5.8_9_8_4, -4.8_3_2_0, -6.8_0_8_6, -6.5_3_9_1, -5.6_1_7_2, -5.5_8_2_0, -5.5_3_5_2, 1.7_8_8_1, 3.6_2_8_9, -6.5_1_1_7, -3.4_7_8_5, -9.5_0_0_0, -6.0_3_5_2, -6.8_1_2_5, -6.0_1_9_5, -6.6_8_3_6, -5.4_7_2_7, -6.2_8_1_2, -6.0_3_9_1, -7.3_3_9_8, -7.4_2_9_7, -7.4_8_4_4, -6.5_8_2_0, -5.8_7_8_9, -5.5_3_1_2] )
# fmt: on
torch.testing.assert_close(out[0, 0, :30], lowerCAmelCase, atol=1e-5, rtol=1e-5 )
@unittest.skip('''Model is curently gated''' )
@slow
def lowercase__ ( self ):
"""simple docstring"""
lowerCamelCase_ ='''Simply put, the theory of relativity states that 1) the laws of physics are the same everywhere in the universe and 2) the passage of time and the length of objects can vary depending on the observer\'s frame of reference.\n\nThe first part of the theory, that the laws of physics are the same everywhere, is known as the "princi'''
lowerCamelCase_ ='''Simply put, the theory of relativity states that '''
lowerCamelCase_ =LlamaTokenizer.from_pretrained('''meta-llama/Llama-2-13b-chat-hf''' )
lowerCamelCase_ =tokenizer.encode(lowerCAmelCase, return_tensors='''pt''' )
lowerCamelCase_ =LlamaForCausalLM.from_pretrained(
'''meta-llama/Llama-2-13b-chat-hf''', device_map='''sequential''', use_safetensors=lowerCAmelCase )
# greedy generation outputs
lowerCamelCase_ =model.generate(lowerCAmelCase, max_new_tokens=64, top_p=lowerCAmelCase, temperature=1, do_sample=lowerCAmelCase )
lowerCamelCase_ =tokenizer.decode(generated_ids[0], skip_special_tokens=lowerCAmelCase )
self.assertEqual(lowerCAmelCase, lowerCAmelCase )
| 75 |
"""simple docstring"""
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A (unittest.TestCase):
'''simple docstring'''
def __init__( self : List[Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Tuple=13 , UpperCAmelCase_ : List[Any]=7 , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=99 , UpperCAmelCase_ : str=32 , UpperCAmelCase_ : Tuple=5 , UpperCAmelCase_ : Union[str, Any]=4 , UpperCAmelCase_ : Any=37 , UpperCAmelCase_ : int="gelu" , UpperCAmelCase_ : Any=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Dict=512 , UpperCAmelCase_ : Optional[Any]=16 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : str=0.02 , UpperCAmelCase_ : str=4 , ) ->Tuple:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_attention_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = type_vocab_size
snake_case_ = type_sequence_label_size
snake_case_ = initializer_range
snake_case_ = num_choices
def lowerCAmelCase ( self : Optional[int] ) ->str:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_attention_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = None
if self.use_token_type_ids:
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=UpperCAmelCase_ , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def lowerCAmelCase ( self : List[str] ) ->Dict:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """token_type_ids""": token_type_ids, """attention_mask""": attention_mask}
return config, inputs_dict
@require_flax
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Union[str, Any] = True
__lowercase: int = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = FlaxRoFormerModelTester(self )
@slow
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""junnyu/roformer_chinese_small""" , from_pt=UpperCAmelCase_ )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
@require_flax
class __A (unittest.TestCase):
'''simple docstring'''
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = FlaxRoFormerForMaskedLM.from_pretrained("""junnyu/roformer_chinese_base""" )
snake_case_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
snake_case_ = model(UpperCAmelCase_ )[0]
snake_case_ = 50_000
snake_case_ = (1, 6, vocab_size)
self.assertEqual(output.shape , UpperCAmelCase_ )
snake_case_ = jnp.array(
[[[-0.1_205, -1.0_265, 0.2_922], [-1.5_134, 0.1_974, 0.1_519], [-5.0_135, -3.9_003, -0.8_404]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=1E-4 ) )
| 347 | 0 |
import os
from pickle import UnpicklingError
from typing import Dict, Tuple
import jax
import jax.numpy as jnp
import numpy as np
from flax.serialization import from_bytes
from flax.traverse_util import flatten_dict, unflatten_dict
import transformers
from .utils import logging
a_ = logging.get_logger(__name__)
def lowerCamelCase__ ( _a , _a , _a , _a=False):
try:
import torch # noqa: F401
except ImportError:
logger.error(
"Loading a PyTorch model in Flax, requires both PyTorch and Flax to be installed. Please see"
" https://pytorch.org/ and https://flax.readthedocs.io/en/latest/installation.html for installation"
" instructions.")
raise
if not is_sharded:
SCREAMING_SNAKE_CASE : str = os.path.abspath(_a)
logger.info(f"Loading PyTorch weights from {pt_path}")
SCREAMING_SNAKE_CASE : Union[str, Any] = torch.load(_a , map_location="cpu")
logger.info(f"PyTorch checkpoint contains {sum(t.numel() for t in pt_state_dict.values()):,} parameters.")
SCREAMING_SNAKE_CASE : str = convert_pytorch_state_dict_to_flax(_a , _a)
else:
# model is sharded and pytorch_checkpoint_path already contains the list of .pt shard files
SCREAMING_SNAKE_CASE : List[str] = convert_pytorch_sharded_state_dict_to_flax(_a , _a)
return flax_state_dict
def lowerCamelCase__ ( _a , _a , _a , _a , ):
def is_key_or_prefix_key_in_dict(_a) -> bool:
return len(set(_a) & {key, (model_prefix,) + key}) > 0
# layer norm
SCREAMING_SNAKE_CASE : Dict = pt_tuple_key[:-1] + ("scale",)
if pt_tuple_key[-1] in ["weight", "gamma"] and is_key_or_prefix_key_in_dict(_a):
return renamed_pt_tuple_key, pt_tensor
# batch norm layer mean
SCREAMING_SNAKE_CASE : Tuple = pt_tuple_key[:-1] + ("mean",)
if pt_tuple_key[-1] == "running_mean" and not is_key_or_prefix_key_in_dict(_a):
return renamed_pt_tuple_key, pt_tensor
# batch norm layer var
SCREAMING_SNAKE_CASE : List[str] = pt_tuple_key[:-1] + ("var",)
if pt_tuple_key[-1] == "running_var" and not is_key_or_prefix_key_in_dict(_a):
return renamed_pt_tuple_key, pt_tensor
# embedding
SCREAMING_SNAKE_CASE : int = pt_tuple_key[:-1] + ("embedding",)
if pt_tuple_key[-1] == "weight" and is_key_or_prefix_key_in_dict(_a):
return renamed_pt_tuple_key, pt_tensor
# conv layer
SCREAMING_SNAKE_CASE : Any = pt_tuple_key[:-1] + ("kernel",)
if pt_tuple_key[-1] == "weight" and pt_tensor.ndim == 4 and not is_key_or_prefix_key_in_dict(_a):
SCREAMING_SNAKE_CASE : Optional[int] = pt_tensor.transpose(2 , 3 , 1 , 0)
return renamed_pt_tuple_key, pt_tensor
# linear layer
SCREAMING_SNAKE_CASE : Union[str, Any] = pt_tuple_key[:-1] + ("kernel",)
if pt_tuple_key[-1] == "weight" and not is_key_or_prefix_key_in_dict(_a):
SCREAMING_SNAKE_CASE : Tuple = pt_tensor.T
return renamed_pt_tuple_key, pt_tensor
# old PyTorch layer norm weight
SCREAMING_SNAKE_CASE : Dict = pt_tuple_key[:-1] + ("weight",)
if pt_tuple_key[-1] == "gamma":
return renamed_pt_tuple_key, pt_tensor
# old PyTorch layer norm bias
SCREAMING_SNAKE_CASE : List[str] = pt_tuple_key[:-1] + ("bias",)
if pt_tuple_key[-1] == "beta":
return renamed_pt_tuple_key, pt_tensor
# New `weight_norm` from https://github.com/huggingface/transformers/pull/24030
SCREAMING_SNAKE_CASE : Optional[int] = None
if pt_tuple_key[-3::2] == ("parametrizations", "original0"):
SCREAMING_SNAKE_CASE : List[str] = pt_tuple_key[-2] + "_g"
elif pt_tuple_key[-3::2] == ("parametrizations", "original1"):
SCREAMING_SNAKE_CASE : Any = pt_tuple_key[-2] + "_v"
if name is not None:
SCREAMING_SNAKE_CASE : str = pt_tuple_key[:-3] + (name,)
return renamed_pt_tuple_key, pt_tensor
return pt_tuple_key, pt_tensor
def lowerCamelCase__ ( _a , _a):
# convert pytorch tensor to numpy
SCREAMING_SNAKE_CASE : int = {k: v.numpy() for k, v in pt_state_dict.items()}
SCREAMING_SNAKE_CASE : List[Any] = flax_model.base_model_prefix
# use params dict if the model contains batch norm layers
if "params" in flax_model.params:
SCREAMING_SNAKE_CASE : Dict = flax_model.params["params"]
else:
SCREAMING_SNAKE_CASE : Union[str, Any] = flax_model.params
SCREAMING_SNAKE_CASE : Optional[Any] = flatten_dict(_a)
# add batch_stats keys,values to dict
if "batch_stats" in flax_model.params:
SCREAMING_SNAKE_CASE : int = flatten_dict(flax_model.params["batch_stats"])
random_flax_state_dict.update(_a)
SCREAMING_SNAKE_CASE : Dict = {}
SCREAMING_SNAKE_CASE : int = (model_prefix not in flax_model_params) and (
model_prefix in {k.split(".")[0] for k in pt_state_dict.keys()}
)
SCREAMING_SNAKE_CASE : Union[str, Any] = (model_prefix in flax_model_params) and (
model_prefix not in {k.split(".")[0] for k in pt_state_dict.keys()}
)
# Need to change some parameters name to match Flax names
for pt_key, pt_tensor in pt_state_dict.items():
SCREAMING_SNAKE_CASE : Optional[int] = tuple(pt_key.split("."))
# remove base model prefix if necessary
SCREAMING_SNAKE_CASE : Dict = pt_tuple_key[0] == model_prefix
if load_model_with_head_into_base_model and has_base_model_prefix:
SCREAMING_SNAKE_CASE : Dict = pt_tuple_key[1:]
# Correctly rename weight parameters
SCREAMING_SNAKE_CASE ,SCREAMING_SNAKE_CASE : int = rename_key_and_reshape_tensor(
_a , _a , _a , _a)
# add model prefix if necessary
SCREAMING_SNAKE_CASE : Optional[Any] = (model_prefix,) + flax_key in random_flax_state_dict
if load_base_model_into_model_with_head and require_base_model_prefix:
SCREAMING_SNAKE_CASE : Any = (model_prefix,) + flax_key
if flax_key in random_flax_state_dict:
if flax_tensor.shape != random_flax_state_dict[flax_key].shape:
raise ValueError(
f"PyTorch checkpoint seems to be incorrect. Weight {pt_key} was expected to be of shape "
f"{random_flax_state_dict[flax_key].shape}, but is {flax_tensor.shape}.")
# add batch stats if the model contains batchnorm layers
if "batch_stats" in flax_model.params:
if "mean" in flax_key[-1] or "var" in flax_key[-1]:
SCREAMING_SNAKE_CASE : Dict = jnp.asarray(_a)
continue
# remove num_batches_tracked key
if "num_batches_tracked" in flax_key[-1]:
flax_state_dict.pop(_a , _a)
continue
# also add unexpected weight so that warning is thrown
SCREAMING_SNAKE_CASE : List[Any] = jnp.asarray(_a)
else:
# also add unexpected weight so that warning is thrown
SCREAMING_SNAKE_CASE : str = jnp.asarray(_a)
return unflatten_dict(_a)
def lowerCamelCase__ ( _a , _a):
import torch
# Load the index
SCREAMING_SNAKE_CASE : int = {}
for shard_file in shard_filenames:
# load using msgpack utils
SCREAMING_SNAKE_CASE : Union[str, Any] = torch.load(_a)
SCREAMING_SNAKE_CASE : Tuple = {k: v.numpy() for k, v in pt_state_dict.items()}
SCREAMING_SNAKE_CASE : Tuple = flax_model.base_model_prefix
# use params dict if the model contains batch norm layers and then add batch_stats keys,values to dict
if "batch_stats" in flax_model.params:
SCREAMING_SNAKE_CASE : str = flax_model.params["params"]
SCREAMING_SNAKE_CASE : Union[str, Any] = flatten_dict(_a)
random_flax_state_dict.update(flatten_dict(flax_model.params["batch_stats"]))
else:
SCREAMING_SNAKE_CASE : Any = flax_model.params
SCREAMING_SNAKE_CASE : Tuple = flatten_dict(_a)
SCREAMING_SNAKE_CASE : List[Any] = (model_prefix not in flax_model_params) and (
model_prefix in {k.split(".")[0] for k in pt_state_dict.keys()}
)
SCREAMING_SNAKE_CASE : Union[str, Any] = (model_prefix in flax_model_params) and (
model_prefix not in {k.split(".")[0] for k in pt_state_dict.keys()}
)
# Need to change some parameters name to match Flax names
for pt_key, pt_tensor in pt_state_dict.items():
SCREAMING_SNAKE_CASE : int = tuple(pt_key.split("."))
# remove base model prefix if necessary
SCREAMING_SNAKE_CASE : Optional[int] = pt_tuple_key[0] == model_prefix
if load_model_with_head_into_base_model and has_base_model_prefix:
SCREAMING_SNAKE_CASE : Tuple = pt_tuple_key[1:]
# Correctly rename weight parameters
SCREAMING_SNAKE_CASE ,SCREAMING_SNAKE_CASE : Optional[int] = rename_key_and_reshape_tensor(
_a , _a , _a , _a)
# add model prefix if necessary
SCREAMING_SNAKE_CASE : Optional[int] = (model_prefix,) + flax_key in random_flax_state_dict
if load_base_model_into_model_with_head and require_base_model_prefix:
SCREAMING_SNAKE_CASE : Optional[Any] = (model_prefix,) + flax_key
if flax_key in random_flax_state_dict:
if flax_tensor.shape != random_flax_state_dict[flax_key].shape:
raise ValueError(
f"PyTorch checkpoint seems to be incorrect. Weight {pt_key} was expected to be of shape "
f"{random_flax_state_dict[flax_key].shape}, but is {flax_tensor.shape}.")
# add batch stats if the model contains batchnorm layers
if "batch_stats" in flax_model.params:
if "mean" in flax_key[-1]:
SCREAMING_SNAKE_CASE : Dict = jnp.asarray(_a)
continue
if "var" in flax_key[-1]:
SCREAMING_SNAKE_CASE : Tuple = jnp.asarray(_a)
continue
# remove num_batches_tracked key
if "num_batches_tracked" in flax_key[-1]:
flax_state_dict.pop(_a , _a)
continue
# also add unexpected weight so that warning is thrown
SCREAMING_SNAKE_CASE : int = jnp.asarray(_a)
else:
# also add unexpected weight so that warning is thrown
SCREAMING_SNAKE_CASE : str = jnp.asarray(_a)
return unflatten_dict(_a)
def lowerCamelCase__ ( _a , _a):
SCREAMING_SNAKE_CASE : str = os.path.abspath(_a)
logger.info(f"Loading Flax weights from {flax_checkpoint_path}")
# import correct flax class
SCREAMING_SNAKE_CASE : int = getattr(_a , "Flax" + model.__class__.__name__)
# load flax weight dict
with open(_a , "rb") as state_f:
try:
SCREAMING_SNAKE_CASE : str = from_bytes(_a , state_f.read())
except UnpicklingError:
raise EnvironmentError(f"Unable to convert {flax_checkpoint_path} to Flax deserializable object. ")
return load_flax_weights_in_pytorch_model(_a , _a)
def lowerCamelCase__ ( _a , _a):
try:
import torch # noqa: F401
except ImportError:
logger.error(
"Loading a Flax weights in PyTorch, requires both PyTorch and Flax to be installed. Please see"
" https://pytorch.org/ and https://flax.readthedocs.io/en/latest/installation.html for installation"
" instructions.")
raise
# check if we have bf16 weights
SCREAMING_SNAKE_CASE : int = flatten_dict(jax.tree_util.tree_map(lambda _a: x.dtype == jnp.bfloataa , _a)).values()
if any(_a):
# convert all weights to fp32 if the are bf16 since torch.from_numpy can-not handle bf16
# and bf16 is not fully supported in PT yet.
logger.warning(
"Found ``bfloat16`` weights in Flax model. Casting all ``bfloat16`` weights to ``float32`` "
"before loading those in PyTorch model.")
SCREAMING_SNAKE_CASE : Optional[Any] = jax.tree_util.tree_map(
lambda _a: params.astype(np.floataa) if params.dtype == jnp.bfloataa else params , _a)
SCREAMING_SNAKE_CASE : Optional[Any] = flatten_dict(_a)
SCREAMING_SNAKE_CASE : Optional[int] = pt_model.state_dict()
SCREAMING_SNAKE_CASE : Union[str, Any] = (pt_model.base_model_prefix in flax_state) and (
pt_model.base_model_prefix not in {k.split(".")[0] for k in pt_model_dict.keys()}
)
SCREAMING_SNAKE_CASE : Dict = (pt_model.base_model_prefix not in flax_state) and (
pt_model.base_model_prefix in {k.split(".")[0] for k in pt_model_dict.keys()}
)
# keep track of unexpected & missing keys
SCREAMING_SNAKE_CASE : int = []
SCREAMING_SNAKE_CASE : int = set(pt_model_dict.keys())
for flax_key_tuple, flax_tensor in flax_state_dict.items():
SCREAMING_SNAKE_CASE : Dict = flax_key_tuple[0] == pt_model.base_model_prefix
SCREAMING_SNAKE_CASE : Union[str, Any] = ".".join((pt_model.base_model_prefix,) + flax_key_tuple) in pt_model_dict
# adapt flax_key to prepare for loading from/to base model only
if load_model_with_head_into_base_model and has_base_model_prefix:
SCREAMING_SNAKE_CASE : Union[str, Any] = flax_key_tuple[1:]
elif load_base_model_into_model_with_head and require_base_model_prefix:
SCREAMING_SNAKE_CASE : Tuple = (pt_model.base_model_prefix,) + flax_key_tuple
# rename flax weights to PyTorch format
if flax_key_tuple[-1] == "kernel" and flax_tensor.ndim == 4 and ".".join(_a) not in pt_model_dict:
# conv layer
SCREAMING_SNAKE_CASE : List[str] = flax_key_tuple[:-1] + ("weight",)
SCREAMING_SNAKE_CASE : Tuple = jnp.transpose(_a , (3, 2, 0, 1))
elif flax_key_tuple[-1] == "kernel" and ".".join(_a) not in pt_model_dict:
# linear layer
SCREAMING_SNAKE_CASE : int = flax_key_tuple[:-1] + ("weight",)
SCREAMING_SNAKE_CASE : Any = flax_tensor.T
elif flax_key_tuple[-1] in ["scale", "embedding"]:
SCREAMING_SNAKE_CASE : Union[str, Any] = flax_key_tuple[:-1] + ("weight",)
# adding batch stats from flax batch norm to pt
elif "mean" in flax_key_tuple[-1]:
SCREAMING_SNAKE_CASE : Union[str, Any] = flax_key_tuple[:-1] + ("running_mean",)
elif "var" in flax_key_tuple[-1]:
SCREAMING_SNAKE_CASE : Union[str, Any] = flax_key_tuple[:-1] + ("running_var",)
if "batch_stats" in flax_state:
SCREAMING_SNAKE_CASE : Dict = ".".join(flax_key_tuple[1:]) # Remove the params/batch_stats header
else:
SCREAMING_SNAKE_CASE : List[str] = ".".join(_a)
# We also need to look at `pt_model_dict` and see if there are keys requiring further transformation.
SCREAMING_SNAKE_CASE : Optional[int] = {}
# New `weight_norm` from https://github.com/huggingface/transformers/pull/24030
for key in pt_model_dict:
SCREAMING_SNAKE_CASE : int = key.split(".")
SCREAMING_SNAKE_CASE : int = None
if key_components[-3::2] == ["parametrizations", "original0"]:
SCREAMING_SNAKE_CASE : Union[str, Any] = key_components[-2] + "_g"
elif key_components[-3::2] == ["parametrizations", "original1"]:
SCREAMING_SNAKE_CASE : Dict = key_components[-2] + "_v"
if name is not None:
SCREAMING_SNAKE_CASE : Optional[int] = key_components[:-3] + [name]
SCREAMING_SNAKE_CASE : List[str] = ".".join(_a)
SCREAMING_SNAKE_CASE : str = key
if flax_key in special_pt_names:
SCREAMING_SNAKE_CASE : Optional[Any] = special_pt_names[flax_key]
if flax_key in pt_model_dict:
if flax_tensor.shape != pt_model_dict[flax_key].shape:
raise ValueError(
f"Flax checkpoint seems to be incorrect. Weight {flax_key_tuple} was expected "
f"to be of shape {pt_model_dict[flax_key].shape}, but is {flax_tensor.shape}.")
else:
# add weight to pytorch dict
SCREAMING_SNAKE_CASE : Dict = np.asarray(_a) if not isinstance(_a , np.ndarray) else flax_tensor
SCREAMING_SNAKE_CASE : int = torch.from_numpy(_a)
# remove from missing keys
missing_keys.remove(_a)
else:
# weight is not expected by PyTorch model
unexpected_keys.append(_a)
pt_model.load_state_dict(_a)
# re-transform missing_keys to list
SCREAMING_SNAKE_CASE : str = list(_a)
if len(_a) > 0:
logger.warning(
"Some weights of the Flax model were not used when initializing the PyTorch model"
f" {pt_model.__class__.__name__}: {unexpected_keys}\n- This IS expected if you are initializing"
f" {pt_model.__class__.__name__} from a Flax model trained on another task or with another architecture"
" (e.g. initializing a BertForSequenceClassification model from a FlaxBertForPreTraining model).\n- This"
f" IS NOT expected if you are initializing {pt_model.__class__.__name__} from a Flax model that you expect"
" to be exactly identical (e.g. initializing a BertForSequenceClassification model from a"
" FlaxBertForSequenceClassification model).")
else:
logger.warning(f"All Flax model weights were used when initializing {pt_model.__class__.__name__}.\n")
if len(_a) > 0:
logger.warning(
f"Some weights of {pt_model.__class__.__name__} were not initialized from the Flax model and are newly"
f" initialized: {missing_keys}\nYou should probably TRAIN this model on a down-stream task to be able to"
" use it for predictions and inference.")
else:
logger.warning(
f"All the weights of {pt_model.__class__.__name__} were initialized from the Flax model.\n"
"If your task is similar to the task the model of the checkpoint was trained on, "
f"you can already use {pt_model.__class__.__name__} for predictions without further training.")
return pt_model | 76 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> bool:
snake_case_ = get_failure_array(_SCREAMING_SNAKE_CASE )
# 2) Step through text searching for pattern
snake_case_ , snake_case_ = 0, 0 # index into text, pattern
while i < len(_SCREAMING_SNAKE_CASE ):
if pattern[j] == text[i]:
if j == (len(_SCREAMING_SNAKE_CASE ) - 1):
return True
j += 1
# if this is a prefix in our pattern
# just go back far enough to continue
elif j > 0:
snake_case_ = failure[j - 1]
continue
i += 1
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> list[int]:
snake_case_ = [0]
snake_case_ = 0
snake_case_ = 1
while j < len(_SCREAMING_SNAKE_CASE ):
if pattern[i] == pattern[j]:
i += 1
elif i > 0:
snake_case_ = failure[i - 1]
continue
j += 1
failure.append(_SCREAMING_SNAKE_CASE )
return failure
if __name__ == "__main__":
# Test 1)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abc1abc12'
__SCREAMING_SNAKE_CASE : Optional[int] = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
__SCREAMING_SNAKE_CASE : List[str] = 'alskfjaldsk23adsfabcabc'
assert kmp(pattern, texta) and not kmp(pattern, texta)
# Test 2)
__SCREAMING_SNAKE_CASE : int = 'ABABX'
__SCREAMING_SNAKE_CASE : Optional[Any] = 'ABABZABABYABABX'
assert kmp(pattern, text)
# Test 3)
__SCREAMING_SNAKE_CASE : Any = 'AAAB'
__SCREAMING_SNAKE_CASE : List[Any] = 'ABAAAAAB'
assert kmp(pattern, text)
# Test 4)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abcdabcy'
__SCREAMING_SNAKE_CASE : str = 'abcxabcdabxabcdabcdabcy'
assert kmp(pattern, text)
# Test 5)
__SCREAMING_SNAKE_CASE : Any = 'aabaabaaa'
assert get_failure_array(pattern) == [0, 1, 0, 1, 2, 3, 4, 5, 2]
| 347 | 0 |
"""simple docstring"""
# Copyright 2023 The HuggingFace Team. All rights reserved.
#
# 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.
from typing import TYPE_CHECKING
# rely on isort to merge the imports
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_tokenizers_available, is_torch_available
_UpperCamelCase : Optional[Any] = {"configuration_mra": ["MRA_PRETRAINED_CONFIG_ARCHIVE_MAP", "MraConfig"]}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
_UpperCamelCase : Any = [
"MRA_PRETRAINED_MODEL_ARCHIVE_LIST",
"MraForMaskedLM",
"MraForMultipleChoice",
"MraForQuestionAnswering",
"MraForSequenceClassification",
"MraForTokenClassification",
"MraLayer",
"MraModel",
"MraPreTrainedModel",
]
if TYPE_CHECKING:
from .configuration_mra import MRA_PRETRAINED_CONFIG_ARCHIVE_MAP, MraConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_mra import (
MRA_PRETRAINED_MODEL_ARCHIVE_LIST,
MraForMaskedLM,
MraForMultipleChoice,
MraForQuestionAnswering,
MraForSequenceClassification,
MraForTokenClassification,
MraLayer,
MraModel,
MraPreTrainedModel,
)
else:
import sys
_UpperCamelCase : List[str] = _LazyModule(__name__, globals()["__file__"], _import_structure)
| 77 |
"""simple docstring"""
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __A (snake_case__):
'''simple docstring'''
@slow
@require_torch
def lowerCAmelCase ( self : Union[str, Any] ) ->Dict:
"""simple docstring"""
snake_case_ = EncoderDecoderModel.from_encoder_decoder_pretrained("""prajjwal1/bert-tiny""" , """prajjwal1/bert-tiny""" )
snake_case_ = BertTokenizer.from_pretrained("""bert-base-uncased""" )
snake_case_ = bertabert.config.encoder.vocab_size
snake_case_ = tokenizer.sep_token_id
snake_case_ = tokenizer.cls_token_id
snake_case_ = 128
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""train[:1%]""" )
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""validation[:1%]""" )
snake_case_ = train_dataset.select(range(32 ) )
snake_case_ = val_dataset.select(range(16 ) )
snake_case_ = 4
def _map_to_encoder_decoder_inputs(UpperCAmelCase_ : int ):
# Tokenizer will automatically set [BOS] <text> [EOS]
snake_case_ = tokenizer(batch["""article"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=512 )
snake_case_ = tokenizer(batch["""highlights"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=128 )
snake_case_ = inputs.input_ids
snake_case_ = inputs.attention_mask
snake_case_ = outputs.input_ids
snake_case_ = outputs.input_ids.copy()
snake_case_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch["""labels"""]
]
snake_case_ = outputs.attention_mask
assert all(len(UpperCAmelCase_ ) == 512 for x in inputs.input_ids )
assert all(len(UpperCAmelCase_ ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(UpperCAmelCase_ : Union[str, Any] ):
snake_case_ = pred.label_ids
snake_case_ = pred.predictions
# all unnecessary tokens are removed
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(UpperCAmelCase_ ) )] ) / len(UpperCAmelCase_ )
return {"accuracy": accuracy}
# map train dataset
snake_case_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
train_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
# same for validation dataset
snake_case_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
val_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
snake_case_ = self.get_auto_remove_tmp_dir()
snake_case_ = SeqaSeqTrainingArguments(
output_dir=UpperCAmelCase_ , per_device_train_batch_size=UpperCAmelCase_ , per_device_eval_batch_size=UpperCAmelCase_ , predict_with_generate=UpperCAmelCase_ , evaluation_strategy="""steps""" , do_train=UpperCAmelCase_ , do_eval=UpperCAmelCase_ , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
snake_case_ = SeqaSeqTrainer(
model=UpperCAmelCase_ , args=UpperCAmelCase_ , compute_metrics=_compute_metrics , train_dataset=UpperCAmelCase_ , eval_dataset=UpperCAmelCase_ , tokenizer=UpperCAmelCase_ , )
# start training
trainer.train()
| 347 | 0 |
"""simple docstring"""
import argparse
import json
import os
import fairseq
import torch
from fairseq.data import Dictionary
# Register SEW's fairseq modules
from sew_asapp import tasks # noqa: F401
from transformers import (
SEWConfig,
SEWForCTC,
SEWModel,
WavaVecaCTCTokenizer,
WavaVecaFeatureExtractor,
WavaVecaProcessor,
logging,
)
logging.set_verbosity_info()
snake_case_ = logging.get_logger(__name__)
snake_case_ = {
"""post_extract_proj""": """feature_projection""",
"""encoder.pos_conv.0""": """encoder.pos_conv_embed.conv""",
"""self_attn.k_proj""": """encoder.layers.*.attention.k_proj""",
"""self_attn.v_proj""": """encoder.layers.*.attention.v_proj""",
"""self_attn.q_proj""": """encoder.layers.*.attention.q_proj""",
"""self_attn.out_proj""": """encoder.layers.*.attention.out_proj""",
"""self_attn_layer_norm""": """encoder.layers.*.layer_norm""",
"""fc1""": """encoder.layers.*.feed_forward.intermediate_dense""",
"""fc2""": """encoder.layers.*.feed_forward.output_dense""",
"""final_layer_norm""": """encoder.layers.*.final_layer_norm""",
"""encoder.upsample.0""": """encoder.upsample.projection""",
"""encoder.layer_norm""": """encoder.layer_norm""",
"""w2v_model.layer_norm""": """layer_norm""",
"""w2v_encoder.proj""": """lm_head""",
"""mask_emb""": """masked_spec_embed""",
}
def _lowerCAmelCase ( lowercase_ , lowercase_ , lowercase_ , lowercase_ , lowercase_ ):
for attribute in key.split('.' ):
UpperCAmelCase = getattr(lowercase_ , lowercase_ )
if weight_type is not None:
UpperCAmelCase = getattr(lowercase_ , lowercase_ ).shape
else:
UpperCAmelCase = hf_pointer.shape
assert hf_shape == value.shape, (
F"""Shape of hf {key + '.' + weight_type if weight_type is not None else ''} is {hf_shape}, but should be"""
F""" {value.shape} for {full_name}"""
)
if weight_type == "weight":
UpperCAmelCase = value
elif weight_type == "weight_g":
UpperCAmelCase = value
elif weight_type == "weight_v":
UpperCAmelCase = value
elif weight_type == "bias":
UpperCAmelCase = value
else:
UpperCAmelCase = value
logger.info(F"""{key + '.' + weight_type if weight_type is not None else ''} was initialized from {full_name}.""" )
def _lowerCAmelCase ( lowercase_ , lowercase_ , lowercase_ ):
UpperCAmelCase = []
UpperCAmelCase = fairseq_model.state_dict()
UpperCAmelCase = hf_model.sew.feature_extractor if is_finetuned else hf_model.feature_extractor
for name, value in fairseq_dict.items():
UpperCAmelCase = False
if "conv_layers" in name:
load_conv_layer(
lowercase_ , lowercase_ , lowercase_ , lowercase_ , hf_model.config.feat_extract_norm == 'group' , )
UpperCAmelCase = True
else:
for key, mapped_key in MAPPING.items():
UpperCAmelCase = 'sew.' + mapped_key if (is_finetuned and mapped_key != 'lm_head') else mapped_key
if key in name or key.split('w2v_model.' )[-1] == name.split('.' )[0]:
UpperCAmelCase = True
if "*" in mapped_key:
UpperCAmelCase = name.split(lowercase_ )[0].split('.' )[-2]
UpperCAmelCase = mapped_key.replace('*' , lowercase_ )
if "weight_g" in name:
UpperCAmelCase = 'weight_g'
elif "weight_v" in name:
UpperCAmelCase = 'weight_v'
elif "weight" in name:
UpperCAmelCase = 'weight'
elif "bias" in name:
UpperCAmelCase = 'bias'
else:
UpperCAmelCase = None
set_recursively(lowercase_ , lowercase_ , lowercase_ , lowercase_ , lowercase_ )
continue
if not is_used:
unused_weights.append(lowercase_ )
logger.warning(F"""Unused weights: {unused_weights}""" )
def _lowerCAmelCase ( lowercase_ , lowercase_ , lowercase_ , lowercase_ , lowercase_ ):
UpperCAmelCase = full_name.split('conv_layers.' )[-1]
UpperCAmelCase = name.split('.' )
UpperCAmelCase = int(items[0] )
UpperCAmelCase = int(items[1] )
if type_id == 0:
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.bias.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found."""
)
UpperCAmelCase = value
logger.info(F"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].conv.weight.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found."""
)
UpperCAmelCase = value
logger.info(F"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape, (
F"""{full_name} has size {value.shape}, but {feature_extractor[layer_id].layer_norm.bias.data.shape} was"""
" found."
)
UpperCAmelCase = value
logger.info(F"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
assert value.shape == feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape, (
F"""{full_name} has size {value.shape}, but"""
F""" {feature_extractor[layer_id].layer_norm.weight.data.shape} was found."""
)
UpperCAmelCase = value
logger.info(F"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(lowercase_ )
def _lowerCAmelCase ( lowercase_ , lowercase_ ):
UpperCAmelCase = SEWConfig()
if is_finetuned:
UpperCAmelCase = model.wav_encoder.wav_model.cfg
else:
UpperCAmelCase = model.cfg
UpperCAmelCase = fs_config.conv_bias
UpperCAmelCase = eval(fs_config.conv_feature_layers )
UpperCAmelCase = [x[0] for x in conv_layers]
UpperCAmelCase = [x[1] for x in conv_layers]
UpperCAmelCase = [x[2] for x in conv_layers]
UpperCAmelCase = 'gelu'
UpperCAmelCase = 'layer' if fs_config.extractor_mode == 'layer_norm' else 'group'
UpperCAmelCase = 0.0
UpperCAmelCase = fs_config.activation_fn.name
UpperCAmelCase = fs_config.encoder_embed_dim
UpperCAmelCase = 0.0_2
UpperCAmelCase = fs_config.encoder_ffn_embed_dim
UpperCAmelCase = 1e-5
UpperCAmelCase = fs_config.encoder_layerdrop
UpperCAmelCase = fs_config.encoder_attention_heads
UpperCAmelCase = fs_config.conv_pos_groups
UpperCAmelCase = fs_config.conv_pos
UpperCAmelCase = len(lowercase_ )
UpperCAmelCase = fs_config.encoder_layers
UpperCAmelCase = fs_config.squeeze_factor
# take care of any params that are overridden by the Wav2VecCtc model
if is_finetuned:
UpperCAmelCase = model.cfg
UpperCAmelCase = fs_config.final_dropout
UpperCAmelCase = fs_config.layerdrop
UpperCAmelCase = fs_config.activation_dropout
UpperCAmelCase = fs_config.mask_prob > 0 or fs_config.mask_channel_prob > 0
UpperCAmelCase = fs_config.attention_dropout
UpperCAmelCase = fs_config.dropout_input
UpperCAmelCase = fs_config.dropout
UpperCAmelCase = fs_config.mask_channel_length
UpperCAmelCase = fs_config.mask_channel_prob
UpperCAmelCase = fs_config.mask_length
UpperCAmelCase = fs_config.mask_prob
UpperCAmelCase = 'Wav2Vec2FeatureExtractor'
UpperCAmelCase = 'Wav2Vec2CTCTokenizer'
return config
@torch.no_grad()
def _lowerCAmelCase ( lowercase_ , lowercase_ , lowercase_=None , lowercase_=None , lowercase_=True ):
if is_finetuned:
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={'data': '/'.join(dict_path.split('/' )[:-1] )} )
else:
UpperCAmelCase , UpperCAmelCase , UpperCAmelCase = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] )
if config_path is not None:
UpperCAmelCase = SEWConfig.from_pretrained(lowercase_ )
else:
UpperCAmelCase = convert_config(model[0] , lowercase_ )
UpperCAmelCase = model[0].eval()
UpperCAmelCase = True if config.feat_extract_norm == 'layer' else False
UpperCAmelCase = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16000 , padding_value=0 , do_normalize=lowercase_ , return_attention_mask=lowercase_ , )
if is_finetuned:
if dict_path:
UpperCAmelCase = Dictionary.load(lowercase_ )
# important change bos & pad token id since CTC symbol is <pad> and
# not <s> as in fairseq
UpperCAmelCase = target_dict.pad_index
UpperCAmelCase = target_dict.bos_index
UpperCAmelCase = target_dict.pad_index
UpperCAmelCase = target_dict.bos_index
UpperCAmelCase = target_dict.eos_index
UpperCAmelCase = len(target_dict.symbols )
UpperCAmelCase = os.path.join(lowercase_ , 'vocab.json' )
if not os.path.isdir(lowercase_ ):
logger.error('--pytorch_dump_folder_path ({}) should be a directory'.format(lowercase_ ) )
return
os.makedirs(lowercase_ , exist_ok=lowercase_ )
with open(lowercase_ , 'w' , encoding='utf-8' ) as vocab_handle:
json.dump(target_dict.indices , lowercase_ )
UpperCAmelCase = WavaVecaCTCTokenizer(
lowercase_ , unk_token=target_dict.unk_word , pad_token=target_dict.pad_word , bos_token=target_dict.bos_word , eos_token=target_dict.eos_word , word_delimiter_token='|' , do_lower_case=lowercase_ , )
UpperCAmelCase = WavaVecaProcessor(feature_extractor=lowercase_ , tokenizer=lowercase_ )
processor.save_pretrained(lowercase_ )
UpperCAmelCase = SEWForCTC(lowercase_ )
else:
UpperCAmelCase = SEWModel(lowercase_ )
feature_extractor.save_pretrained(lowercase_ )
recursively_load_weights(lowercase_ , lowercase_ , lowercase_ )
hf_model.save_pretrained(lowercase_ )
if __name__ == "__main__":
snake_case_ = argparse.ArgumentParser()
parser.add_argument("""--pytorch_dump_folder_path""", default=None, type=str, help="""Path to the output PyTorch model.""")
parser.add_argument("""--checkpoint_path""", default=None, type=str, help="""Path to fairseq checkpoint""")
parser.add_argument("""--dict_path""", default=None, type=str, help="""Path to dict of fine-tuned model""")
parser.add_argument("""--config_path""", default=None, type=str, help="""Path to hf config.json of model to convert""")
parser.add_argument(
"""--is_finetuned""", action="""store_true""", help="""Whether the model to convert is a fine-tuned model or not"""
)
snake_case_ = parser.parse_args()
convert_sew_checkpoint(
args.checkpoint_path, args.pytorch_dump_folder_path, args.config_path, args.dict_path, args.is_finetuned
)
| 78 |
"""simple docstring"""
# this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.:
# python ./utils/get_modified_files.py utils src tests examples
#
# it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered
# since the output of this script is fed into Makefile commands it doesn't print a newline after the results
import re
import subprocess
import sys
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8')
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split()
__SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:])
__SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""")
__SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)]
print(' '.join(relevant_modified_files), end='')
| 347 | 0 |
'''simple docstring'''
import gc
import random
import unittest
import numpy as np
import torch
from transformers import XLMRobertaTokenizer
from diffusers import (
AltDiffusionImgaImgPipeline,
AutoencoderKL,
PNDMScheduler,
UNetaDConditionModel,
)
from diffusers.image_processor import VaeImageProcessor
from diffusers.pipelines.alt_diffusion.modeling_roberta_series import (
RobertaSeriesConfig,
RobertaSeriesModelWithTransformation,
)
from diffusers.utils import floats_tensor, load_image, load_numpy, slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
enable_full_determinism()
class _UpperCAmelCase ( unittest.TestCase ):
"""simple docstring"""
def lowerCAmelCase ( self : int ):
'''simple docstring'''
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
@property
def lowerCAmelCase ( self : Dict ):
'''simple docstring'''
_A = 1
_A = 3
_A = (32, 32)
_A = floats_tensor((batch_size, num_channels) + sizes , rng=random.Random(0 ) ).to(__UpperCAmelCase )
return image
@property
def lowerCAmelCase ( self : str ):
'''simple docstring'''
torch.manual_seed(0 )
_A = UNetaDConditionModel(
block_out_channels=(32, 64) , layers_per_block=2 , sample_size=32 , in_channels=4 , out_channels=4 , down_block_types=("DownBlock2D", "CrossAttnDownBlock2D") , up_block_types=("CrossAttnUpBlock2D", "UpBlock2D") , cross_attention_dim=32 , )
return model
@property
def lowerCAmelCase ( self : List[str] ):
'''simple docstring'''
torch.manual_seed(0 )
_A = AutoencoderKL(
block_out_channels=[32, 64] , in_channels=3 , out_channels=3 , down_block_types=["DownEncoderBlock2D", "DownEncoderBlock2D"] , up_block_types=["UpDecoderBlock2D", "UpDecoderBlock2D"] , latent_channels=4 , )
return model
@property
def lowerCAmelCase ( self : Any ):
'''simple docstring'''
torch.manual_seed(0 )
_A = RobertaSeriesConfig(
hidden_size=32 , project_dim=32 , intermediate_size=37 , layer_norm_eps=1E-05 , num_attention_heads=4 , num_hidden_layers=5 , pad_token_id=1 , vocab_size=5006 , )
return RobertaSeriesModelWithTransformation(__UpperCAmelCase )
@property
def lowerCAmelCase ( self : Optional[int] ):
'''simple docstring'''
def extract(*__UpperCAmelCase : str , **__UpperCAmelCase : List[Any] ):
class _UpperCAmelCase :
"""simple docstring"""
def __init__( self : str ):
'''simple docstring'''
_A = torch.ones([0] )
def lowerCAmelCase ( self : str , __UpperCAmelCase : List[Any] ):
'''simple docstring'''
self.pixel_values.to(__UpperCAmelCase )
return self
return Out()
return extract
def lowerCAmelCase ( self : List[str] ):
'''simple docstring'''
_A = "cpu" # ensure determinism for the device-dependent torch.Generator
_A = self.dummy_cond_unet
_A = PNDMScheduler(skip_prk_steps=__UpperCAmelCase )
_A = self.dummy_vae
_A = self.dummy_text_encoder
_A = XLMRobertaTokenizer.from_pretrained("hf-internal-testing/tiny-xlm-roberta" )
_A = 77
_A = self.dummy_image.to(__UpperCAmelCase )
_A = init_image / 2 + 0.5
# make sure here that pndm scheduler skips prk
_A = AltDiffusionImgaImgPipeline(
unet=__UpperCAmelCase , scheduler=__UpperCAmelCase , vae=__UpperCAmelCase , text_encoder=__UpperCAmelCase , tokenizer=__UpperCAmelCase , safety_checker=__UpperCAmelCase , feature_extractor=self.dummy_extractor , )
_A = VaeImageProcessor(vae_scale_factor=alt_pipe.vae_scale_factor , do_normalize=__UpperCAmelCase )
_A = alt_pipe.to(__UpperCAmelCase )
alt_pipe.set_progress_bar_config(disable=__UpperCAmelCase )
_A = "A painting of a squirrel eating a burger"
_A = torch.Generator(device=__UpperCAmelCase ).manual_seed(0 )
_A = alt_pipe(
[prompt] , generator=__UpperCAmelCase , guidance_scale=6.0 , num_inference_steps=2 , output_type="np" , image=__UpperCAmelCase , )
_A = output.images
_A = torch.Generator(device=__UpperCAmelCase ).manual_seed(0 )
_A = alt_pipe(
[prompt] , generator=__UpperCAmelCase , guidance_scale=6.0 , num_inference_steps=2 , output_type="np" , image=__UpperCAmelCase , return_dict=__UpperCAmelCase , )[0]
_A = image[0, -3:, -3:, -1]
_A = image_from_tuple[0, -3:, -3:, -1]
assert image.shape == (1, 32, 32, 3)
_A = np.array([0.4427, 0.3731, 0.4249, 0.4941, 0.4546, 0.4148, 0.4193, 0.4666, 0.4499] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 5E-3
assert np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() < 5E-3
@unittest.skipIf(torch_device != "cuda" , "This test requires a GPU" )
def lowerCAmelCase ( self : Optional[Any] ):
'''simple docstring'''
_A = self.dummy_cond_unet
_A = PNDMScheduler(skip_prk_steps=__UpperCAmelCase )
_A = self.dummy_vae
_A = self.dummy_text_encoder
_A = XLMRobertaTokenizer.from_pretrained("hf-internal-testing/tiny-xlm-roberta" )
_A = 77
_A = self.dummy_image.to(__UpperCAmelCase )
# put models in fp16
_A = unet.half()
_A = vae.half()
_A = bert.half()
# make sure here that pndm scheduler skips prk
_A = AltDiffusionImgaImgPipeline(
unet=__UpperCAmelCase , scheduler=__UpperCAmelCase , vae=__UpperCAmelCase , text_encoder=__UpperCAmelCase , tokenizer=__UpperCAmelCase , safety_checker=__UpperCAmelCase , feature_extractor=self.dummy_extractor , )
_A = VaeImageProcessor(vae_scale_factor=alt_pipe.vae_scale_factor , do_normalize=__UpperCAmelCase )
_A = alt_pipe.to(__UpperCAmelCase )
alt_pipe.set_progress_bar_config(disable=__UpperCAmelCase )
_A = "A painting of a squirrel eating a burger"
_A = torch.manual_seed(0 )
_A = alt_pipe(
[prompt] , generator=__UpperCAmelCase , num_inference_steps=2 , output_type="np" , image=__UpperCAmelCase , ).images
assert image.shape == (1, 32, 32, 3)
@unittest.skipIf(torch_device != "cuda" , "This test requires a GPU" )
def lowerCAmelCase ( self : Union[str, Any] ):
'''simple docstring'''
_A = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
"/img2img/sketch-mountains-input.jpg" )
# resize to resolution that is divisible by 8 but not 16 or 32
_A = init_image.resize((760, 504) )
_A = "BAAI/AltDiffusion"
_A = AltDiffusionImgaImgPipeline.from_pretrained(
__UpperCAmelCase , safety_checker=__UpperCAmelCase , )
pipe.to(__UpperCAmelCase )
pipe.set_progress_bar_config(disable=__UpperCAmelCase )
pipe.enable_attention_slicing()
_A = "A fantasy landscape, trending on artstation"
_A = torch.manual_seed(0 )
_A = pipe(
prompt=__UpperCAmelCase , image=__UpperCAmelCase , strength=0.75 , guidance_scale=7.5 , generator=__UpperCAmelCase , output_type="np" , )
_A = output.images[0]
_A = image[255:258, 383:386, -1]
assert image.shape == (504, 760, 3)
_A = np.array([0.9358, 0.9397, 0.9599, 0.9901, 1.0000, 1.0000, 0.9882, 1.0000, 1.0000] )
assert np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
@slow
@require_torch_gpu
class _UpperCAmelCase ( unittest.TestCase ):
"""simple docstring"""
def lowerCAmelCase ( self : int ):
'''simple docstring'''
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def lowerCAmelCase ( self : int ):
'''simple docstring'''
_A = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
"/img2img/sketch-mountains-input.jpg" )
_A = init_image.resize((768, 512) )
_A = load_numpy(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/img2img/fantasy_landscape_alt.npy" )
_A = "BAAI/AltDiffusion"
_A = AltDiffusionImgaImgPipeline.from_pretrained(
__UpperCAmelCase , safety_checker=__UpperCAmelCase , )
pipe.to(__UpperCAmelCase )
pipe.set_progress_bar_config(disable=__UpperCAmelCase )
pipe.enable_attention_slicing()
_A = "A fantasy landscape, trending on artstation"
_A = torch.manual_seed(0 )
_A = pipe(
prompt=__UpperCAmelCase , image=__UpperCAmelCase , strength=0.75 , guidance_scale=7.5 , generator=__UpperCAmelCase , output_type="np" , )
_A = output.images[0]
assert image.shape == (512, 768, 3)
# img2img is flaky across GPUs even in fp32, so using MAE here
assert np.abs(expected_image - image ).max() < 1E-2
| 79 |
"""simple docstring"""
import argparse
import json
import os
import fairseq
import torch
from fairseq.data import Dictionary
from transformers import (
WavaVecaConfig,
WavaVecaCTCTokenizer,
WavaVecaFeatureExtractor,
WavaVecaForCTC,
WavaVecaForPreTraining,
WavaVecaProcessor,
logging,
)
from transformers.models.wavaveca.modeling_wavaveca import WavaVecaForSequenceClassification
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'post_extract_proj': 'feature_projection.projection',
'encoder.pos_conv.0': 'encoder.pos_conv_embed.conv',
'self_attn.k_proj': 'encoder.layers.*.attention.k_proj',
'self_attn.v_proj': 'encoder.layers.*.attention.v_proj',
'self_attn.q_proj': 'encoder.layers.*.attention.q_proj',
'self_attn.out_proj': 'encoder.layers.*.attention.out_proj',
'self_attn_layer_norm': 'encoder.layers.*.layer_norm',
'fc1': 'encoder.layers.*.feed_forward.intermediate_dense',
'fc2': 'encoder.layers.*.feed_forward.output_dense',
'final_layer_norm': 'encoder.layers.*.final_layer_norm',
'encoder.layer_norm': 'encoder.layer_norm',
'adapter_layer': 'encoder.layers.*.adapter_layer',
'w2v_model.layer_norm': 'feature_projection.layer_norm',
'quantizer.weight_proj': 'quantizer.weight_proj',
'quantizer.vars': 'quantizer.codevectors',
'project_q': 'project_q',
'final_proj': 'project_hid',
'w2v_encoder.proj': 'lm_head',
'mask_emb': 'masked_spec_embed',
'pooling_layer.linear': 'projector',
'pooling_layer.projection': 'classifier',
}
__SCREAMING_SNAKE_CASE : List[Any] = [
'lm_head',
'quantizer.weight_proj',
'quantizer.codevectors',
'project_q',
'project_hid',
'projector',
'classifier',
]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = {}
with open(_SCREAMING_SNAKE_CASE , """r""" ) as file:
for line_number, line in enumerate(_SCREAMING_SNAKE_CASE ):
snake_case_ = line.strip()
if line:
snake_case_ = line.split()
snake_case_ = line_number
snake_case_ = words[0]
snake_case_ = value
return result
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
for attribute in key.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ).shape
elif weight_type is not None and weight_type == "param":
snake_case_ = hf_pointer
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = shape_pointer.shape
# let's reduce dimension
snake_case_ = value[0]
else:
snake_case_ = hf_pointer.shape
if hf_shape != value.shape:
raise ValueError(
f"""Shape of hf {key + "." + weight_type if weight_type is not None else ""} is {hf_shape}, but should be"""
f""" {value.shape} for {full_name}""" )
if weight_type == "weight":
snake_case_ = value
elif weight_type == "weight_g":
snake_case_ = value
elif weight_type == "weight_v":
snake_case_ = value
elif weight_type == "bias":
snake_case_ = value
elif weight_type == "param":
for attribute in hf_param_name.split(""".""" ):
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = value
else:
snake_case_ = value
logger.info(f"""{key + "." + weight_type if weight_type is not None else ""} was initialized from {full_name}.""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Tuple:
snake_case_ = None
for param_key in PARAM_MAPPING.keys():
if full_name.endswith(_SCREAMING_SNAKE_CASE ):
snake_case_ = PARAM_MAPPING[full_name.split(""".""" )[-1]]
snake_case_ = """param"""
if weight_type is not None and weight_type != "param":
snake_case_ = """.""".join([key, weight_type] )
elif weight_type is not None and weight_type == "param":
snake_case_ = """.""".join([key, hf_param_name] )
else:
snake_case_ = key
snake_case_ = value if """lm_head""" in full_key else value[0]
__SCREAMING_SNAKE_CASE : int = {
'W_a': 'linear_1.weight',
'W_b': 'linear_2.weight',
'b_a': 'linear_1.bias',
'b_b': 'linear_2.bias',
'ln_W': 'norm.weight',
'ln_b': 'norm.bias',
}
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> List[str]:
snake_case_ = False
for key, mapped_key in MAPPING.items():
snake_case_ = """wav2vec2.""" + mapped_key if mapped_key not in TOP_LEVEL_KEYS else mapped_key
if key in name or key.split("""w2v_model.""" )[-1] == name.split(""".""" )[0]:
snake_case_ = True
if "*" in mapped_key:
snake_case_ = name.split(_SCREAMING_SNAKE_CASE )[0].split(""".""" )[-2]
snake_case_ = mapped_key.replace("""*""" , _SCREAMING_SNAKE_CASE )
if "weight_g" in name:
snake_case_ = """weight_g"""
elif "weight_v" in name:
snake_case_ = """weight_v"""
elif "bias" in name:
snake_case_ = """bias"""
elif "weight" in name:
# TODO: don't match quantizer.weight_proj
snake_case_ = """weight"""
else:
snake_case_ = None
if hf_dict is not None:
rename_dict(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
else:
set_recursively(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
return is_used
return is_used
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = []
snake_case_ = fairseq_model.state_dict()
snake_case_ = hf_model.wavaveca.feature_extractor
for name, value in fairseq_dict.items():
snake_case_ = False
if "conv_layers" in name:
load_conv_layer(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , hf_model.config.feat_extract_norm == """group""" , )
snake_case_ = True
else:
snake_case_ = load_wavaveca_layer(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not is_used:
unused_weights.append(_SCREAMING_SNAKE_CASE )
logger.warning(f"""Unused weights: {unused_weights}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
snake_case_ = full_name.split("""conv_layers.""" )[-1]
snake_case_ = name.split(""".""" )
snake_case_ = int(items[0] )
snake_case_ = int(items[1] )
if type_id == 0:
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].conv.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].conv.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract conv layer {layer_id} was initialized from {full_name}.""" )
elif (type_id == 2 and not use_group_norm) or (type_id == 2 and layer_id == 0 and use_group_norm):
if "bias" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.bias.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
elif "weight" in name:
if value.shape != feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape:
raise ValueError(
f"""{full_name} has size {value.shape}, but"""
f""" {feature_extractor.conv_layers[layer_id].layer_norm.weight.data.shape} was found.""" )
snake_case_ = value
logger.info(f"""Feat extract layer norm weight of layer {layer_id} was initialized from {full_name}.""" )
else:
unused_weights.append(_SCREAMING_SNAKE_CASE )
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=True , _SCREAMING_SNAKE_CASE=False ) -> int:
if config_path is not None:
snake_case_ = WavaVecaConfig.from_pretrained(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaConfig()
if is_seq_class:
snake_case_ = read_txt_into_dict(_SCREAMING_SNAKE_CASE )
snake_case_ = idalabel
snake_case_ = WavaVecaForSequenceClassification(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
feature_extractor.save_pretrained(_SCREAMING_SNAKE_CASE )
elif is_finetuned:
if dict_path:
snake_case_ = Dictionary.load(_SCREAMING_SNAKE_CASE )
# important change bos & pad token id since CTC symbol is <pad> and
# not <s> as in fairseq
snake_case_ = target_dict.pad_index
snake_case_ = target_dict.bos_index
snake_case_ = target_dict.eos_index
snake_case_ = len(target_dict.symbols )
snake_case_ = os.path.join(_SCREAMING_SNAKE_CASE , """vocab.json""" )
if not os.path.isdir(_SCREAMING_SNAKE_CASE ):
logger.error("""--pytorch_dump_folder_path ({}) should be a directory""".format(_SCREAMING_SNAKE_CASE ) )
return
os.makedirs(_SCREAMING_SNAKE_CASE , exist_ok=_SCREAMING_SNAKE_CASE )
snake_case_ = target_dict.indices
# fairseq has the <pad> and <s> switched
snake_case_ = 0
snake_case_ = 1
with open(_SCREAMING_SNAKE_CASE , """w""" , encoding="""utf-8""" ) as vocab_handle:
json.dump(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaCTCTokenizer(
_SCREAMING_SNAKE_CASE , unk_token=target_dict.unk_word , pad_token=target_dict.pad_word , bos_token=target_dict.bos_word , eos_token=target_dict.eos_word , word_delimiter_token="""|""" , do_lower_case=_SCREAMING_SNAKE_CASE , )
snake_case_ = True if config.feat_extract_norm == """layer""" else False
snake_case_ = WavaVecaFeatureExtractor(
feature_size=1 , sampling_rate=16_000 , padding_value=0 , do_normalize=_SCREAMING_SNAKE_CASE , return_attention_mask=_SCREAMING_SNAKE_CASE , )
snake_case_ = WavaVecaProcessor(feature_extractor=_SCREAMING_SNAKE_CASE , tokenizer=_SCREAMING_SNAKE_CASE )
processor.save_pretrained(_SCREAMING_SNAKE_CASE )
snake_case_ = WavaVecaForCTC(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = WavaVecaForPreTraining(_SCREAMING_SNAKE_CASE )
if is_finetuned or is_seq_class:
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task(
[checkpoint_path] , arg_overrides={"""data""": """/""".join(dict_path.split("""/""" )[:-1] )} )
else:
snake_case_ = argparse.Namespace(task="""audio_pretraining""" )
snake_case_ = fairseq.tasks.setup_task(_SCREAMING_SNAKE_CASE )
snake_case_ , snake_case_ , snake_case_ = fairseq.checkpoint_utils.load_model_ensemble_and_task([checkpoint_path] , task=_SCREAMING_SNAKE_CASE )
snake_case_ = model[0].eval()
recursively_load_weights(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , not is_finetuned )
hf_wavavec.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : str = argparse.ArgumentParser()
parser.add_argument('--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model.')
parser.add_argument('--checkpoint_path', default=None, type=str, help='Path to fairseq checkpoint')
parser.add_argument('--dict_path', default=None, type=str, help='Path to dict of fine-tuned model')
parser.add_argument('--config_path', default=None, type=str, help='Path to hf config.json of model to convert')
parser.add_argument(
'--not_finetuned', action='store_true', help='Whether the model to convert is a fine-tuned model or not'
)
parser.add_argument(
'--is_seq_class',
action='store_true',
help='Whether the model to convert is a fine-tuned sequence classification model or not',
)
__SCREAMING_SNAKE_CASE : Any = parser.parse_args()
__SCREAMING_SNAKE_CASE : List[Any] = not args.not_finetuned and not args.is_seq_class
convert_wavaveca_checkpoint(
args.checkpoint_path,
args.pytorch_dump_folder_path,
args.config_path,
args.dict_path,
is_finetuned,
args.is_seq_class,
)
| 347 | 0 |
'''simple docstring'''
from typing import Dict, List, Optional, Union
import numpy as np
from ...image_processing_utils import BaseImageProcessor, BatchFeature, get_size_dict
from ...image_transforms import rescale, resize, to_channel_dimension_format
from ...image_utils import (
ChannelDimension,
ImageInput,
PILImageResampling,
make_list_of_images,
to_numpy_array,
valid_images,
)
from ...utils import TensorType, is_vision_available, logging
if is_vision_available():
import PIL
a__ : Optional[int] = logging.get_logger(__name__)
def _UpperCamelCase ( __A , __A ) -> str:
'''simple docstring'''
UpperCamelCase__ = b.T
UpperCamelCase__ = np.sum(np.square(__A ) , axis=1 )
UpperCamelCase__ = np.sum(np.square(__A ) , axis=0 )
UpperCamelCase__ = np.matmul(__A , __A )
UpperCamelCase__ = aa[:, None] - 2 * ab + ba[None, :]
return d
def _UpperCamelCase ( __A , __A ) -> Dict:
'''simple docstring'''
UpperCamelCase__ = x.reshape(-1 , 3 )
UpperCamelCase__ = squared_euclidean_distance(__A , __A )
return np.argmin(__A , axis=1 )
class lowercase_ ( a__ ):
__UpperCAmelCase = ['pixel_values']
def __init__( self , a = None , a = True , a = None , a = PILImageResampling.BILINEAR , a = True , a = True , **a , ):
super().__init__(**a )
UpperCamelCase__ = size if size is not None else {"height": 2_56, "width": 2_56}
UpperCamelCase__ = get_size_dict(a )
UpperCamelCase__ = np.array(a ) if clusters is not None else None
UpperCamelCase__ = do_resize
UpperCamelCase__ = size
UpperCamelCase__ = resample
UpperCamelCase__ = do_normalize
UpperCamelCase__ = do_color_quantize
def __a ( self , a , a , a = PILImageResampling.BILINEAR , a = None , **a , ):
UpperCamelCase__ = get_size_dict(a )
if "height" not in size or "width" not in size:
raise ValueError(f'''Size dictionary must contain both height and width keys. Got {size.keys()}''' )
return resize(
a , size=(size["height"], size["width"]) , resample=a , data_format=a , **a )
def __a ( self , a , a = None , ):
UpperCamelCase__ = rescale(image=a , scale=1 / 127.5 , data_format=a )
UpperCamelCase__ = image - 1
return image
def __a ( self , a , a = None , a = None , a = None , a = None , a = None , a = None , a = None , a = ChannelDimension.FIRST , **a , ):
UpperCamelCase__ = do_resize if do_resize is not None else self.do_resize
UpperCamelCase__ = size if size is not None else self.size
UpperCamelCase__ = get_size_dict(a )
UpperCamelCase__ = resample if resample is not None else self.resample
UpperCamelCase__ = do_normalize if do_normalize is not None else self.do_normalize
UpperCamelCase__ = do_color_quantize if do_color_quantize is not None else self.do_color_quantize
UpperCamelCase__ = clusters if clusters is not None else self.clusters
UpperCamelCase__ = np.array(a )
UpperCamelCase__ = make_list_of_images(a )
if not valid_images(a ):
raise ValueError(
"Invalid image type. Must be of type PIL.Image.Image, numpy.ndarray, "
"torch.Tensor, tf.Tensor or jax.ndarray." )
if do_resize and size is None or resample is None:
raise ValueError("Size and resample must be specified if do_resize is True." )
if do_color_quantize and clusters is None:
raise ValueError("Clusters must be specified if do_color_quantize is True." )
# All transformations expect numpy arrays.
UpperCamelCase__ = [to_numpy_array(a ) for image in images]
if do_resize:
UpperCamelCase__ = [self.resize(image=a , size=a , resample=a ) for image in images]
if do_normalize:
UpperCamelCase__ = [self.normalize(image=a ) for image in images]
if do_color_quantize:
UpperCamelCase__ = [to_channel_dimension_format(a , ChannelDimension.LAST ) for image in images]
# color quantize from (batch_size, height, width, 3) to (batch_size, height, width)
UpperCamelCase__ = np.array(a )
UpperCamelCase__ = color_quantize(a , a ).reshape(images.shape[:-1] )
# flatten to (batch_size, height*width)
UpperCamelCase__ = images.shape[0]
UpperCamelCase__ = images.reshape(a , -1 )
# We need to convert back to a list of images to keep consistent behaviour across processors.
UpperCamelCase__ = list(a )
else:
UpperCamelCase__ = [to_channel_dimension_format(a , a ) for image in images]
UpperCamelCase__ = {"input_ids": images}
return BatchFeature(data=a , tensor_type=a )
| 80 |
"""simple docstring"""
import tempfile
import unittest
import numpy as np
import transformers
from transformers import GPTaTokenizer, GPTJConfig, is_flax_available, is_torch_available
from transformers.testing_utils import is_pt_flax_cross_test, require_flax, tooslow
from ...generation.test_flax_utils import FlaxGenerationTesterMixin
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax
import jax.numpy as jnp
from transformers.modeling_flax_pytorch_utils import (
convert_pytorch_state_dict_to_flax,
load_flax_weights_in_pytorch_model,
)
from transformers.models.gptj.modeling_flax_gptj import FlaxGPTJForCausalLM, FlaxGPTJModel
if is_torch_available():
import torch
class __A :
'''simple docstring'''
def __init__( self : Dict , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Any=14 , UpperCAmelCase_ : Union[str, Any]=7 , UpperCAmelCase_ : Tuple=True , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Union[str, Any]=False , UpperCAmelCase_ : Union[str, Any]=True , UpperCAmelCase_ : str=99 , UpperCAmelCase_ : Union[str, Any]=32 , UpperCAmelCase_ : List[Any]=4 , UpperCAmelCase_ : Optional[int]=4 , UpperCAmelCase_ : int=4 , UpperCAmelCase_ : str=37 , UpperCAmelCase_ : Any="gelu" , UpperCAmelCase_ : str=0.1 , UpperCAmelCase_ : Union[str, Any]=0.1 , UpperCAmelCase_ : int=512 , UpperCAmelCase_ : Tuple=0.02 , ) ->List[str]:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_input_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = rotary_dim
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = initializer_range
snake_case_ = None
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
snake_case_ = vocab_size - 1
def lowerCAmelCase ( self : int ) ->Optional[int]:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_input_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = GPTJConfig(
vocab_size=self.vocab_size , n_embd=self.hidden_size , n_layer=self.num_hidden_layers , n_head=self.num_attention_heads , n_positions=self.max_position_embeddings , use_cache=UpperCAmelCase_ , bos_token_id=self.bos_token_id , eos_token_id=self.eos_token_id , pad_token_id=self.pad_token_id , rotary_dim=self.rotary_dim , )
return (config, input_ids, input_mask)
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """attention_mask""": attention_mask}
return config, inputs_dict
def lowerCAmelCase ( self : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : List[Any] , UpperCAmelCase_ : List[str] , UpperCAmelCase_ : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.ones((input_ids.shape[0], max_decoder_length) , dtype="""i4""" )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , attention_mask=UpperCAmelCase_ , past_key_values=outputs_cache.past_key_values , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
def lowerCAmelCase ( self : Union[str, Any] , UpperCAmelCase_ : int , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = 20
snake_case_ = model_class_name(UpperCAmelCase_ )
snake_case_ = jnp.concatenate(
[attention_mask, jnp.zeros((attention_mask.shape[0], max_decoder_length - attention_mask.shape[1]) )] , axis=-1 , )
snake_case_ = model.init_cache(input_ids.shape[0] , UpperCAmelCase_ )
snake_case_ = jnp.broadcast_to(
jnp.arange(input_ids.shape[-1] - 1 )[None, :] , (input_ids.shape[0], input_ids.shape[-1] - 1) )
snake_case_ = model(
input_ids[:, :-1] , attention_mask=UpperCAmelCase_ , past_key_values=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = jnp.array(input_ids.shape[0] * [[input_ids.shape[-1] - 1]] , dtype="""i4""" )
snake_case_ = model(
input_ids[:, -1:] , past_key_values=outputs_cache.past_key_values , attention_mask=UpperCAmelCase_ , position_ids=UpperCAmelCase_ , )
snake_case_ = model(UpperCAmelCase_ , attention_mask=UpperCAmelCase_ )
snake_case_ = np.max(np.abs((outputs_cache_next[0][:, -1, :5] - outputs[0][:, -1, :5]) ) )
self.parent.assertTrue(diff < 1E-3 , msg=F"""Max diff is {diff}""" )
@require_flax
class __A (snake_case__ , snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Any = (FlaxGPTJModel, FlaxGPTJForCausalLM) if is_flax_available() else ()
__lowercase: List[str] = (FlaxGPTJForCausalLM,) if is_flax_available() else ()
def lowerCAmelCase ( self : Tuple ) ->List[str]:
"""simple docstring"""
snake_case_ = FlaxGPTJModelTester(self )
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward(UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] ) ->Any:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ , snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs()
self.model_tester.check_use_cache_forward_with_attn_mask(
UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ , UpperCAmelCase_ )
@tooslow
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = GPTaTokenizer.from_pretrained("""gpt2""" , pad_token="""<|endoftext|>""" , padding_side="""left""" )
snake_case_ = tokenizer(["""Hello this is a long string""", """Hey"""] , return_tensors="""np""" , padding=UpperCAmelCase_ , truncation=UpperCAmelCase_ )
snake_case_ = FlaxGPTJForCausalLM.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = False
snake_case_ = model.config.eos_token_id
snake_case_ = jax.jit(model.generate )
snake_case_ = jit_generate(
inputs["""input_ids"""] , attention_mask=inputs["""attention_mask"""] , pad_token_id=tokenizer.pad_token_id ).sequences
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = [
"""Hello this is a long string of text.\n\nI'm trying to get the text of the""",
"""Hey, I'm a little late to the party. I'm going to""",
]
self.assertListEqual(UpperCAmelCase_ , UpperCAmelCase_ )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = convert_pytorch_state_dict_to_flax(pt_model.state_dict() , UpperCAmelCase_ )
snake_case_ = fx_state
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
pt_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = model_class.from_pretrained(UpperCAmelCase_ , from_pt=UpperCAmelCase_ )
snake_case_ = fx_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output_loaded, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output_loaded[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@is_pt_flax_cross_test
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
snake_case_ , snake_case_ = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
with self.subTest(model_class.__name__ ):
# prepare inputs
snake_case_ = self._prepare_for_class(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = {k: torch.tensor(v.tolist() ) for k, v in prepared_inputs_dict.items()}
# load corresponding PyTorch class
snake_case_ = model_class.__name__[4:] # Skip the "Flax" at the beginning
snake_case_ = getattr(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = pt_model_class(UpperCAmelCase_ ).eval()
snake_case_ = model_class(UpperCAmelCase_ , dtype=jnp.floataa )
snake_case_ = load_flax_weights_in_pytorch_model(UpperCAmelCase_ , fx_model.params )
snake_case_ , snake_case_ = pt_inputs["""input_ids"""].shape
snake_case_ = np.random.randint(0 , seq_length - 1 , size=(batch_size,) )
for batch_idx, start_index in enumerate(UpperCAmelCase_ ):
snake_case_ = 0
snake_case_ = 1
snake_case_ = 0
snake_case_ = 1
# make sure weights are tied in PyTorch
pt_model.tie_weights()
with torch.no_grad():
snake_case_ = pt_model(**UpperCAmelCase_ ).to_tuple()
snake_case_ = fx_model(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
with tempfile.TemporaryDirectory() as tmpdirname:
fx_model.save_pretrained(UpperCAmelCase_ )
snake_case_ = pt_model_class.from_pretrained(UpperCAmelCase_ , from_flax=UpperCAmelCase_ )
with torch.no_grad():
snake_case_ = pt_model_loaded(**UpperCAmelCase_ ).to_tuple()
self.assertEqual(
len(UpperCAmelCase_ ) , len(UpperCAmelCase_ ) , """Output lengths differ between Flax and PyTorch""" )
for fx_output, pt_output in zip(UpperCAmelCase_ , UpperCAmelCase_ ):
self.assert_almost_equals(fx_output[:, -1] , pt_output[:, -1].numpy() , 4E-2 )
@tooslow
def lowerCAmelCase ( self : List[Any] ) ->str:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""EleutherAI/gpt-j-6B""" )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
"""simple docstring"""
from __future__ import annotations
import math
from collections.abc import Callable
def _A ( lowercase , lowercase , lowercase , lowercase = 1_00 , ):
"""simple docstring"""
a =x_start
a =fnc(lowercase )
a =0.0
for _ in range(lowercase ):
# Approximates curve as a sequence of linear lines and sums their length
a =(x_end - x_start) / steps + xa
a =fnc(lowercase )
length += math.hypot(xa - xa , fxa - fxa )
# Increment step
a =xa
a =fxa
return length
if __name__ == "__main__":
def _A ( lowercase ):
"""simple docstring"""
return math.sin(10 * x )
print("""f(x) = sin(10 * x)""")
print("""The length of the curve from x = -10 to x = 10 is:""")
lowerCamelCase_ : Tuple = 1_0
while i <= 1_0_0_0_0_0:
print(F'With {i} steps: {line_length(f, -1_0, 1_0, i)}')
i *= 1_0 | 81 |
"""simple docstring"""
import copy
from ...configuration_utils import PretrainedConfig
from ...utils import logging
from ..auto.configuration_auto import CONFIG_MAPPING
__SCREAMING_SNAKE_CASE : Any = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """upernet"""
def __init__( self : str , UpperCAmelCase_ : List[str]=None , UpperCAmelCase_ : str=512 , UpperCAmelCase_ : int=0.02 , UpperCAmelCase_ : Optional[Any]=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : Tuple=0.4 , UpperCAmelCase_ : Tuple=384 , UpperCAmelCase_ : Union[str, Any]=256 , UpperCAmelCase_ : str=1 , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : Dict , ) ->Union[str, Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
if backbone_config is None:
logger.info("""`backbone_config` is `None`. Initializing the config with the default `ResNet` backbone.""" )
snake_case_ = CONFIG_MAPPING["""resnet"""](out_features=["""stage1""", """stage2""", """stage3""", """stage4"""] )
elif isinstance(UpperCAmelCase_ , UpperCAmelCase_ ):
snake_case_ = backbone_config.get("""model_type""" )
snake_case_ = CONFIG_MAPPING[backbone_model_type]
snake_case_ = config_class.from_dict(UpperCAmelCase_ )
snake_case_ = backbone_config
snake_case_ = hidden_size
snake_case_ = initializer_range
snake_case_ = pool_scales
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_in_channels
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = loss_ignore_index
def lowerCAmelCase ( self : str ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = copy.deepcopy(self.__dict__ )
snake_case_ = self.backbone_config.to_dict()
snake_case_ = self.__class__.model_type
return output
| 347 | 0 |
import shutil
import tempfile
import unittest
from transformers import SPIECE_UNDERLINE, BatchEncoding, MBartTokenizer, MBartTokenizerFast, is_torch_available
from transformers.testing_utils import (
get_tests_dir,
nested_simplify,
require_sentencepiece,
require_tokenizers,
require_torch,
)
from ...test_tokenization_common import TokenizerTesterMixin
A__ = get_tests_dir("""fixtures/test_sentencepiece.model""")
if is_torch_available():
from transformers.models.mbart.modeling_mbart import shift_tokens_right
A__ = 25_00_04
A__ = 25_00_20
@require_sentencepiece
@require_tokenizers
class __lowerCAmelCase ( lowerCamelCase__ , unittest.TestCase ):
__lowerCamelCase = MBartTokenizer
__lowerCamelCase = MBartTokenizerFast
__lowerCamelCase = True
__lowerCamelCase = True
def snake_case ( self ):
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
_lowerCAmelCase = MBartTokenizer(_snake_case , keep_accents=_snake_case )
tokenizer.save_pretrained(self.tmpdirname )
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = MBartTokenizer(_snake_case , keep_accents=_snake_case )
_lowerCAmelCase = tokenizer.tokenize("""This is a test""" )
self.assertListEqual(_snake_case , ["""▁This""", """▁is""", """▁a""", """▁t""", """est"""] )
self.assertListEqual(
tokenizer.convert_tokens_to_ids(_snake_case ) , [value + tokenizer.fairseq_offset for value in [285, 46, 10, 170, 382]] , )
_lowerCAmelCase = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
_snake_case , [
SPIECE_UNDERLINE + """I""",
SPIECE_UNDERLINE + """was""",
SPIECE_UNDERLINE + """b""",
"""or""",
"""n""",
SPIECE_UNDERLINE + """in""",
SPIECE_UNDERLINE + """""",
"""9""",
"""2""",
"""0""",
"""0""",
"""0""",
""",""",
SPIECE_UNDERLINE + """and""",
SPIECE_UNDERLINE + """this""",
SPIECE_UNDERLINE + """is""",
SPIECE_UNDERLINE + """f""",
"""al""",
"""s""",
"""é""",
""".""",
] , )
_lowerCAmelCase = tokenizer.convert_tokens_to_ids(_snake_case )
self.assertListEqual(
_snake_case , [
value + tokenizer.fairseq_offset
for value in [8, 21, 84, 55, 24, 19, 7, 2, 602, 347, 347, 347, 3, 12, 66, 46, 72, 80, 6, 2, 4]
# ^ unk: 2 + 1 = 3 unk: 2 + 1 = 3 ^
] , )
_lowerCAmelCase = tokenizer.convert_ids_to_tokens(_snake_case )
self.assertListEqual(
_snake_case , [
SPIECE_UNDERLINE + """I""",
SPIECE_UNDERLINE + """was""",
SPIECE_UNDERLINE + """b""",
"""or""",
"""n""",
SPIECE_UNDERLINE + """in""",
SPIECE_UNDERLINE + """""",
"""<unk>""",
"""2""",
"""0""",
"""0""",
"""0""",
""",""",
SPIECE_UNDERLINE + """and""",
SPIECE_UNDERLINE + """this""",
SPIECE_UNDERLINE + """is""",
SPIECE_UNDERLINE + """f""",
"""al""",
"""s""",
"""<unk>""",
""".""",
] , )
def snake_case ( self ):
"""simple docstring"""
if not self.test_slow_tokenizer:
# as we don't have a slow version, we can't compare the outputs between slow and fast versions
return
_lowerCAmelCase = (self.rust_tokenizer_class, """hf-internal-testing/tiny-random-mbart""", {})
for tokenizer, pretrained_name, kwargs in self.tokenizers_list:
with self.subTest(F'{tokenizer.__class__.__name__} ({pretrained_name})' ):
_lowerCAmelCase = self.rust_tokenizer_class.from_pretrained(_snake_case , **_snake_case )
_lowerCAmelCase = self.tokenizer_class.from_pretrained(_snake_case , **_snake_case )
_lowerCAmelCase = tempfile.mkdtemp()
_lowerCAmelCase = tokenizer_r.save_pretrained(_snake_case )
_lowerCAmelCase = tokenizer_p.save_pretrained(_snake_case )
# Checks it save with the same files + the tokenizer.json file for the fast one
self.assertTrue(any("""tokenizer.json""" in f for f in tokenizer_r_files ) )
_lowerCAmelCase = tuple(f for f in tokenizer_r_files if """tokenizer.json""" not in f )
self.assertSequenceEqual(_snake_case , _snake_case )
# Checks everything loads correctly in the same way
_lowerCAmelCase = tokenizer_r.from_pretrained(_snake_case )
_lowerCAmelCase = tokenizer_p.from_pretrained(_snake_case )
# Check special tokens are set accordingly on Rust and Python
for key in tokenizer_pp.special_tokens_map:
self.assertTrue(hasattr(_snake_case , _snake_case ) )
# self.assertEqual(getattr(tokenizer_rp, key), getattr(tokenizer_pp, key))
# self.assertEqual(getattr(tokenizer_rp, key + "_id"), getattr(tokenizer_pp, key + "_id"))
shutil.rmtree(_snake_case )
# Save tokenizer rust, legacy_format=True
_lowerCAmelCase = tempfile.mkdtemp()
_lowerCAmelCase = tokenizer_r.save_pretrained(_snake_case , legacy_format=_snake_case )
_lowerCAmelCase = tokenizer_p.save_pretrained(_snake_case )
# Checks it save with the same files
self.assertSequenceEqual(_snake_case , _snake_case )
# Checks everything loads correctly in the same way
_lowerCAmelCase = tokenizer_r.from_pretrained(_snake_case )
_lowerCAmelCase = tokenizer_p.from_pretrained(_snake_case )
# Check special tokens are set accordingly on Rust and Python
for key in tokenizer_pp.special_tokens_map:
self.assertTrue(hasattr(_snake_case , _snake_case ) )
shutil.rmtree(_snake_case )
# Save tokenizer rust, legacy_format=False
_lowerCAmelCase = tempfile.mkdtemp()
_lowerCAmelCase = tokenizer_r.save_pretrained(_snake_case , legacy_format=_snake_case )
_lowerCAmelCase = tokenizer_p.save_pretrained(_snake_case )
# Checks it saved the tokenizer.json file
self.assertTrue(any("""tokenizer.json""" in f for f in tokenizer_r_files ) )
# Checks everything loads correctly in the same way
_lowerCAmelCase = tokenizer_r.from_pretrained(_snake_case )
_lowerCAmelCase = tokenizer_p.from_pretrained(_snake_case )
# Check special tokens are set accordingly on Rust and Python
for key in tokenizer_pp.special_tokens_map:
self.assertTrue(hasattr(_snake_case , _snake_case ) )
shutil.rmtree(_snake_case )
@require_torch
@require_sentencepiece
@require_tokenizers
class __lowerCAmelCase ( unittest.TestCase ):
__lowerCamelCase = '''facebook/mbart-large-en-ro'''
__lowerCamelCase = [
''' UN Chief Says There Is No Military Solution in Syria''',
''' Secretary-General Ban Ki-moon says his response to Russia\'s stepped up military support for Syria is that "there is no military solution" to the nearly five-year conflict and more weapons will only worsen the violence and misery for millions of people.''',
]
__lowerCamelCase = [
'''Şeful ONU declară că nu există o soluţie militară în Siria''',
'''Secretarul General Ban Ki-moon declară că răspunsul său la intensificarea sprijinului militar al Rusiei'''
''' pentru Siria este că "nu există o soluţie militară" la conflictul de aproape cinci ani şi că noi arme nu vor'''
''' face decât să înrăutăţească violenţele şi mizeria pentru milioane de oameni.''',
]
__lowerCamelCase = [8_274, 127_873, 25_916, 7, 8_622, 2_071, 438, 67_485, 53, 187_895, 23, 51_712, 2, EN_CODE]
@classmethod
def snake_case ( cls ):
"""simple docstring"""
_lowerCAmelCase = MBartTokenizer.from_pretrained(
cls.checkpoint_name , src_lang="""en_XX""" , tgt_lang="""ro_RO""" )
_lowerCAmelCase = 1
return cls
def snake_case ( self ):
"""simple docstring"""
self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""ar_AR"""] , 250001 )
self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""en_EN"""] , 250004 )
self.assertEqual(self.tokenizer.fairseq_tokens_to_ids["""ro_RO"""] , 250020 )
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = self.tokenizer.batch_encode_plus(self.src_text ).input_ids[0]
self.assertListEqual(self.expected_src_tokens , _snake_case )
def snake_case ( self ):
"""simple docstring"""
self.assertIn(_snake_case , self.tokenizer.all_special_ids )
_lowerCAmelCase = [RO_CODE, 884, 9019, 96, 9, 916, 86792, 36, 18743, 15596, 5, 2]
_lowerCAmelCase = self.tokenizer.decode(_snake_case , skip_special_tokens=_snake_case )
_lowerCAmelCase = self.tokenizer.decode(generated_ids[1:] , skip_special_tokens=_snake_case )
self.assertEqual(_snake_case , _snake_case )
self.assertNotIn(self.tokenizer.eos_token , _snake_case )
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = ["""this is gunna be a long sentence """ * 20]
assert isinstance(src_text[0] , _snake_case )
_lowerCAmelCase = 10
_lowerCAmelCase = self.tokenizer(_snake_case , max_length=_snake_case , truncation=_snake_case ).input_ids[0]
self.assertEqual(ids[-2] , 2 )
self.assertEqual(ids[-1] , _snake_case )
self.assertEqual(len(_snake_case ) , _snake_case )
def snake_case ( self ):
"""simple docstring"""
self.assertListEqual(self.tokenizer.convert_tokens_to_ids(["""<mask>""", """ar_AR"""] ) , [250026, 250001] )
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = tempfile.mkdtemp()
_lowerCAmelCase = self.tokenizer.fairseq_tokens_to_ids
self.tokenizer.save_pretrained(_snake_case )
_lowerCAmelCase = MBartTokenizer.from_pretrained(_snake_case )
self.assertDictEqual(new_tok.fairseq_tokens_to_ids , _snake_case )
@require_torch
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = self.tokenizer(self.src_text , text_target=self.tgt_text , padding=_snake_case , return_tensors="""pt""" )
_lowerCAmelCase = shift_tokens_right(batch["""labels"""] , self.tokenizer.pad_token_id )
# fairseq batch: https://gist.github.com/sshleifer/cba08bc2109361a74ac3760a7e30e4f4
assert batch.input_ids[1][-2:].tolist() == [2, EN_CODE]
assert batch.decoder_input_ids[1][0].tolist() == RO_CODE
assert batch.decoder_input_ids[1][-1] == 2
assert batch.labels[1][-2:].tolist() == [2, RO_CODE]
@require_torch
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = self.tokenizer(
self.src_text , text_target=self.tgt_text , padding=_snake_case , truncation=_snake_case , max_length=len(self.expected_src_tokens ) , return_tensors="""pt""" , )
_lowerCAmelCase = shift_tokens_right(batch["""labels"""] , self.tokenizer.pad_token_id )
self.assertIsInstance(_snake_case , _snake_case )
self.assertEqual((2, 14) , batch.input_ids.shape )
self.assertEqual((2, 14) , batch.attention_mask.shape )
_lowerCAmelCase = batch.input_ids.tolist()[0]
self.assertListEqual(self.expected_src_tokens , _snake_case )
self.assertEqual(2 , batch.decoder_input_ids[0, -1] ) # EOS
# Test that special tokens are reset
self.assertEqual(self.tokenizer.prefix_tokens , [] )
self.assertEqual(self.tokenizer.suffix_tokens , [self.tokenizer.eos_token_id, EN_CODE] )
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = self.tokenizer(self.src_text , padding=_snake_case , truncation=_snake_case , max_length=3 , return_tensors="""pt""" )
_lowerCAmelCase = self.tokenizer(
text_target=self.tgt_text , padding=_snake_case , truncation=_snake_case , max_length=10 , return_tensors="""pt""" )
_lowerCAmelCase = targets["""input_ids"""]
_lowerCAmelCase = shift_tokens_right(_snake_case , self.tokenizer.pad_token_id )
self.assertEqual(batch.input_ids.shape[1] , 3 )
self.assertEqual(batch.decoder_input_ids.shape[1] , 10 )
@require_torch
def snake_case ( self ):
"""simple docstring"""
_lowerCAmelCase = self.tokenizer._build_translation_inputs(
"""A test""" , return_tensors="""pt""" , src_lang="""en_XX""" , tgt_lang="""ar_AR""" )
self.assertEqual(
nested_simplify(_snake_case ) , {
# A, test, EOS, en_XX
"""input_ids""": [[62, 3034, 2, 250004]],
"""attention_mask""": [[1, 1, 1, 1]],
# ar_AR
"""forced_bos_token_id""": 250001,
} , )
| 82 |
"""simple docstring"""
import os
import shutil
import tempfile
import unittest
import numpy as np
from transformers import AutoTokenizer, BarkProcessor
from transformers.testing_utils import require_torch, slow
@require_torch
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = """ylacombe/bark-small"""
snake_case_ = tempfile.mkdtemp()
snake_case_ = """en_speaker_1"""
snake_case_ = """This is a test string"""
snake_case_ = """speaker_embeddings_path.json"""
snake_case_ = """speaker_embeddings"""
def lowerCAmelCase ( self : List[str] , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
return AutoTokenizer.from_pretrained(self.checkpoint , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
shutil.rmtree(self.tmpdirname )
def lowerCAmelCase ( self : List[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
processor.save_pretrained(self.tmpdirname )
snake_case_ = BarkProcessor.from_pretrained(self.tmpdirname )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer.get_vocab() )
@slow
def lowerCAmelCase ( self : Dict ) ->int:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
processor.save_pretrained(
self.tmpdirname , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , speaker_embeddings_directory=self.speaker_embeddings_directory , )
snake_case_ = self.get_tokenizer(bos_token="""(BOS)""" , eos_token="""(EOS)""" )
snake_case_ = BarkProcessor.from_pretrained(
self.tmpdirname , self.speaker_embeddings_dict_path , bos_token="""(BOS)""" , eos_token="""(EOS)""" , )
self.assertEqual(processor.tokenizer.get_vocab() , tokenizer_add_kwargs.get_vocab() )
def lowerCAmelCase ( self : Optional[Any] ) ->Any:
"""simple docstring"""
snake_case_ = BarkProcessor.from_pretrained(
pretrained_processor_name_or_path=self.checkpoint , speaker_embeddings_dict_path=self.speaker_embeddings_dict_path , )
snake_case_ = 35
snake_case_ = 2
snake_case_ = 8
snake_case_ = {
"""semantic_prompt""": np.ones(UpperCAmelCase_ ),
"""coarse_prompt""": np.ones((nb_codebooks_coarse, seq_len) ),
"""fine_prompt""": np.ones((nb_codebooks_total, seq_len) ),
}
# test providing already loaded voice_preset
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from npz file
snake_case_ = os.path.join(self.tmpdirname , """file.npz""" )
np.savez(UpperCAmelCase_ , **UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string , voice_preset=UpperCAmelCase_ )
snake_case_ = inputs["""history_prompt"""]
for key in voice_preset:
self.assertListEqual(voice_preset[key].tolist() , processed_voice_preset.get(UpperCAmelCase_ , np.array([] ) ).tolist() )
# test loading voice preset from the hub
snake_case_ = processor(text=self.input_string , voice_preset=self.voice_preset )
def lowerCAmelCase ( self : Tuple ) ->Dict:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = BarkProcessor(tokenizer=UpperCAmelCase_ )
snake_case_ = processor(text=self.input_string )
snake_case_ = tokenizer(
self.input_string , padding="""max_length""" , max_length=256 , add_special_tokens=UpperCAmelCase_ , return_attention_mask=UpperCAmelCase_ , return_token_type_ids=UpperCAmelCase_ , )
for key in encoded_tok.keys():
self.assertListEqual(encoded_tok[key] , encoded_processor[key].squeeze().tolist() )
| 347 | 0 |
'''simple docstring'''
import gc
import random
import unittest
import numpy as np
import torch
from diffusers import DDIMScheduler, KandinskyVaaPipeline, KandinskyVaaPriorPipeline, UNetaDConditionModel, VQModel
from diffusers.utils import floats_tensor, load_numpy, slow, torch_device
from diffusers.utils.testing_utils import enable_full_determinism, require_torch_gpu
from ..test_pipelines_common import PipelineTesterMixin, assert_mean_pixel_difference
enable_full_determinism()
class lowercase__ ( lowercase , unittest.TestCase ):
lowercase__ = KandinskyVaaPipeline
lowercase__ = [
"""image_embeds""",
"""negative_image_embeds""",
]
lowercase__ = ["""image_embeds""", """negative_image_embeds"""]
lowercase__ = [
"""generator""",
"""height""",
"""width""",
"""latents""",
"""guidance_scale""",
"""num_inference_steps""",
"""return_dict""",
"""guidance_scale""",
"""num_images_per_prompt""",
"""output_type""",
"""return_dict""",
]
lowercase__ = False
@property
def UpperCamelCase_ ( self : str ):
'''simple docstring'''
return 32
@property
def UpperCamelCase_ ( self : Tuple ):
'''simple docstring'''
return 32
@property
def UpperCamelCase_ ( self : Tuple ):
'''simple docstring'''
return self.time_input_dim
@property
def UpperCamelCase_ ( self : Optional[Any] ):
'''simple docstring'''
return self.time_input_dim * 4
@property
def UpperCamelCase_ ( self : Tuple ):
'''simple docstring'''
return 100
@property
def UpperCamelCase_ ( self : int ):
'''simple docstring'''
torch.manual_seed(0 )
_UpperCamelCase : Dict = {
'in_channels': 4,
# Out channels is double in channels because predicts mean and variance
'out_channels': 8,
'addition_embed_type': 'image',
'down_block_types': ('ResnetDownsampleBlock2D', 'SimpleCrossAttnDownBlock2D'),
'up_block_types': ('SimpleCrossAttnUpBlock2D', 'ResnetUpsampleBlock2D'),
'mid_block_type': 'UNetMidBlock2DSimpleCrossAttn',
'block_out_channels': (self.block_out_channels_a, self.block_out_channels_a * 2),
'layers_per_block': 1,
'encoder_hid_dim': self.text_embedder_hidden_size,
'encoder_hid_dim_type': 'image_proj',
'cross_attention_dim': self.cross_attention_dim,
'attention_head_dim': 4,
'resnet_time_scale_shift': 'scale_shift',
'class_embed_type': None,
}
_UpperCamelCase : str = UNetaDConditionModel(**lowerCamelCase__ )
return model
@property
def UpperCamelCase_ ( self : str ):
'''simple docstring'''
return {
"block_out_channels": [32, 64],
"down_block_types": ["DownEncoderBlock2D", "AttnDownEncoderBlock2D"],
"in_channels": 3,
"latent_channels": 4,
"layers_per_block": 1,
"norm_num_groups": 8,
"norm_type": "spatial",
"num_vq_embeddings": 12,
"out_channels": 3,
"up_block_types": [
"AttnUpDecoderBlock2D",
"UpDecoderBlock2D",
],
"vq_embed_dim": 4,
}
@property
def UpperCamelCase_ ( self : List[Any] ):
'''simple docstring'''
torch.manual_seed(0 )
_UpperCamelCase : str = VQModel(**self.dummy_movq_kwargs )
return model
def UpperCamelCase_ ( self : Union[str, Any] ):
'''simple docstring'''
_UpperCamelCase : Tuple = self.dummy_unet
_UpperCamelCase : List[Any] = self.dummy_movq
_UpperCamelCase : Optional[Any] = DDIMScheduler(
num_train_timesteps=1000 ,beta_schedule='linear' ,beta_start=0.0_0_0_8_5 ,beta_end=0.0_1_2 ,clip_sample=lowerCamelCase__ ,set_alpha_to_one=lowerCamelCase__ ,steps_offset=1 ,prediction_type='epsilon' ,thresholding=lowerCamelCase__ ,)
_UpperCamelCase : List[str] = {
'unet': unet,
'scheduler': scheduler,
'movq': movq,
}
return components
def UpperCamelCase_ ( self : Any ,lowerCamelCase__ : Dict ,lowerCamelCase__ : str=0 ):
'''simple docstring'''
_UpperCamelCase : Tuple = floats_tensor((1, self.text_embedder_hidden_size) ,rng=random.Random(lowerCamelCase__ ) ).to(lowerCamelCase__ )
_UpperCamelCase : Dict = floats_tensor((1, self.text_embedder_hidden_size) ,rng=random.Random(seed + 1 ) ).to(
lowerCamelCase__ )
if str(lowerCamelCase__ ).startswith('mps' ):
_UpperCamelCase : str = torch.manual_seed(lowerCamelCase__ )
else:
_UpperCamelCase : Tuple = torch.Generator(device=lowerCamelCase__ ).manual_seed(lowerCamelCase__ )
_UpperCamelCase : List[str] = {
'image_embeds': image_embeds,
'negative_image_embeds': negative_image_embeds,
'generator': generator,
'height': 64,
'width': 64,
'guidance_scale': 4.0,
'num_inference_steps': 2,
'output_type': 'np',
}
return inputs
def UpperCamelCase_ ( self : Tuple ):
'''simple docstring'''
_UpperCamelCase : List[Any] = 'cpu'
_UpperCamelCase : Dict = self.get_dummy_components()
_UpperCamelCase : Optional[int] = self.pipeline_class(**lowerCamelCase__ )
_UpperCamelCase : Optional[Any] = pipe.to(lowerCamelCase__ )
pipe.set_progress_bar_config(disable=lowerCamelCase__ )
_UpperCamelCase : Optional[Any] = pipe(**self.get_dummy_inputs(lowerCamelCase__ ) )
_UpperCamelCase : Dict = output.images
_UpperCamelCase : Dict = pipe(
**self.get_dummy_inputs(lowerCamelCase__ ) ,return_dict=lowerCamelCase__ ,)[0]
_UpperCamelCase : Tuple = image[0, -3:, -3:, -1]
_UpperCamelCase : List[Any] = image_from_tuple[0, -3:, -3:, -1]
assert image.shape == (1, 64, 64, 3)
_UpperCamelCase : Any = np.array(
[0.6_2_3_7_9_7_6, 1.0, 0.3_6_4_4_1_3_3_2, 1.0, 0.7_0_6_3_9_6_3_4, 0.2_9_8_7_7_1_8_6, 0.8_5_6_5_2_1_2_5, 0.5_2_1_6_8_4_3, 0.5_4_4_5_4_0_4_6] )
assert (
np.abs(image_slice.flatten() - expected_slice ).max() < 1E-2
), F' expected_slice {expected_slice}, but got {image_slice.flatten()}'
assert (
np.abs(image_from_tuple_slice.flatten() - expected_slice ).max() < 1E-2
), F' expected_slice {expected_slice}, but got {image_from_tuple_slice.flatten()}'
@slow
@require_torch_gpu
class lowercase__ ( unittest.TestCase ):
def UpperCamelCase_ ( self : Optional[Any] ):
'''simple docstring'''
# clean up the VRAM after each test
super().tearDown()
gc.collect()
torch.cuda.empty_cache()
def UpperCamelCase_ ( self : Optional[Any] ):
'''simple docstring'''
_UpperCamelCase : List[str] = load_numpy(
'https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main'
'/kandinskyv22/kandinskyv22_text2img_cat_fp16.npy' )
_UpperCamelCase : Any = KandinskyVaaPriorPipeline.from_pretrained(
'kandinsky-community/kandinsky-2-2-prior' ,torch_dtype=torch.floataa )
pipe_prior.to(lowerCamelCase__ )
_UpperCamelCase : List[str] = KandinskyVaaPipeline.from_pretrained(
'kandinsky-community/kandinsky-2-2-decoder' ,torch_dtype=torch.floataa )
_UpperCamelCase : Union[str, Any] = pipeline.to(lowerCamelCase__ )
pipeline.set_progress_bar_config(disable=lowerCamelCase__ )
_UpperCamelCase : int = 'red cat, 4k photo'
_UpperCamelCase : List[str] = torch.Generator(device='cuda' ).manual_seed(0 )
_UpperCamelCase , _UpperCamelCase : str = pipe_prior(
lowerCamelCase__ ,generator=lowerCamelCase__ ,num_inference_steps=5 ,negative_prompt='' ,).to_tuple()
_UpperCamelCase : Optional[int] = torch.Generator(device='cuda' ).manual_seed(0 )
_UpperCamelCase : Tuple = pipeline(
image_embeds=lowerCamelCase__ ,negative_image_embeds=lowerCamelCase__ ,generator=lowerCamelCase__ ,num_inference_steps=100 ,output_type='np' ,)
_UpperCamelCase : str = output.images[0]
assert image.shape == (512, 512, 3)
assert_mean_pixel_difference(lowerCamelCase__ ,lowerCamelCase__ )
| 83 |
"""simple docstring"""
import argparse
import json
import os
import sys
import tempfile
import unittest
from argparse import Namespace
from dataclasses import dataclass, field
from enum import Enum
from pathlib import Path
from typing import List, Literal, Optional
import yaml
from transformers import HfArgumentParser, TrainingArguments
from transformers.hf_argparser import make_choice_type_function, string_to_bool
# Since Python 3.10, we can use the builtin `|` operator for Union types
# See PEP 604: https://peps.python.org/pep-0604
__SCREAMING_SNAKE_CASE : int = sys.version_info >= (3, 10)
def _a ( _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
return field(default_factory=lambda: default , metadata=_SCREAMING_SNAKE_CASE )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: float
__lowercase: str
__lowercase: bool
@dataclass
class __A :
'''simple docstring'''
__lowercase: int = 42
__lowercase: str = field(default="""toto""" , metadata={"""help""": """help message"""})
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: Optional[bool] = None
class __A (snake_case__):
'''simple docstring'''
__lowercase: str = """titi"""
__lowercase: Any = """toto"""
class __A (snake_case__):
'''simple docstring'''
__lowercase: int = """titi"""
__lowercase: Optional[Any] = """toto"""
__lowercase: List[Any] = 42
@dataclass
class __A :
'''simple docstring'''
__lowercase: BasicEnum = "toto"
def lowerCAmelCase ( self : int ) ->List[Any]:
"""simple docstring"""
snake_case_ = BasicEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: MixedTypeEnum = "toto"
def lowerCAmelCase ( self : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = MixedTypeEnum(self.foo )
@dataclass
class __A :
'''simple docstring'''
__lowercase: Optional[int] = None
__lowercase: Optional[float] = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: Optional[str] = None
__lowercase: Optional[List[str]] = list_field(default=[])
__lowercase: Optional[List[int]] = list_field(default=[])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = list_field(default=[])
__lowercase: List[int] = list_field(default=[1, 2, 3])
__lowercase: List[str] = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
__lowercase: List[float] = list_field(default=[0.1, 0.2, 0.3])
@dataclass
class __A :
'''simple docstring'''
__lowercase: List[int] = field()
__lowercase: str = field()
__lowercase: BasicEnum = field()
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
snake_case_ = BasicEnum(self.required_enum )
@dataclass
class __A :
'''simple docstring'''
__lowercase: int
__lowercase: "BasicEnum" = field()
__lowercase: "Optional[bool]" = None
__lowercase: "str" = field(default="""toto""" , metadata={"""help""": """help message"""})
__lowercase: "List[str]" = list_field(default=["""Hallo""", """Bonjour""", """Hello"""])
if is_python_no_less_than_3_10:
@dataclass
class __A :
'''simple docstring'''
__lowercase: bool = False
__lowercase: bool = True
__lowercase: bool | None = None
@dataclass
class __A :
'''simple docstring'''
__lowercase: int | None = None
__lowercase: float | None = field(default=snake_case__ , metadata={"""help""": """help message"""})
__lowercase: str | None = None
__lowercase: list[str] | None = list_field(default=[])
__lowercase: list[int] | None = list_field(default=[])
class __A (unittest.TestCase):
'''simple docstring'''
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : argparse.ArgumentParser , UpperCAmelCase_ : argparse.ArgumentParser ) ->Optional[int]:
"""simple docstring"""
self.assertEqual(len(a._actions ) , len(b._actions ) )
for x, y in zip(a._actions , b._actions ):
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
snake_case_ = {k: v for k, v in vars(UpperCAmelCase_ ).items() if k != """container"""}
# Choices with mixed type have custom function as "type"
# So we need to compare results directly for equality
if xx.get("""choices""" , UpperCAmelCase_ ) and yy.get("""choices""" , UpperCAmelCase_ ):
for expected_choice in yy["choices"] + xx["choices"]:
self.assertEqual(xx["""type"""](UpperCAmelCase_ ) , yy["""type"""](UpperCAmelCase_ ) )
del xx["type"], yy["type"]
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--bar""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--flag""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = ["""--foo""", """1""", """--baz""", """quux""", """--bar""", """0.5"""]
((snake_case_) , ) = parser.parse_args_into_dataclasses(UpperCAmelCase_ , look_for_args_file=UpperCAmelCase_ )
self.assertFalse(example.flag )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=42 , type=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
expected.add_argument("""--baz""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ , const=UpperCAmelCase_ , nargs="""?""" )
# A boolean no_* argument always has to come after its "default: True" regular counter-part
# and its default must be set to False
expected.add_argument("""--no_baz""" , action="""store_false""" , default=UpperCAmelCase_ , dest="""baz""" )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
snake_case_ = [WithDefaultBoolExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--no_baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """--baz"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """True""", """--baz""", """True""", """--opt""", """True"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
snake_case_ = parser.parse_args(["""--foo""", """False""", """--baz""", """False""", """--opt""", """False"""] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , baz=UpperCAmelCase_ , opt=UpperCAmelCase_ ) )
def lowerCAmelCase ( self : int ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=["""titi""", """toto""", 42] , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args_into_dataclasses([] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.toto )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """titi"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.titi )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
snake_case_ = parser.parse_args_into_dataclasses(["""--foo""", """42"""] )[0]
self.assertEqual(enum_ex.foo , MixedTypeEnum.fourtytwo )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
@dataclass
class __A :
'''simple docstring'''
__lowercase: Literal["titi", "toto", 42] = "toto"
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument(
"""--foo""" , default="""toto""" , choices=("""titi""", """toto""", 42) , type=make_choice_type_function(["""titi""", """toto""", 42] ) , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(args.foo , """toto""" )
snake_case_ = parser.parse_args(["""--foo""", """titi"""] )
self.assertEqual(args.foo , """titi""" )
snake_case_ = parser.parse_args(["""--foo""", """42"""] )
self.assertEqual(args.foo , 42 )
def lowerCAmelCase ( self : Optional[int] ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo_int""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--bar_int""" , nargs="""+""" , default=[1, 2, 3] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
expected.add_argument("""--foo_float""" , nargs="""+""" , default=[0.1, 0.2, 0.3] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(
UpperCAmelCase_ , Namespace(foo_int=[] , bar_int=[1, 2, 3] , foo_str=["""Hallo""", """Bonjour""", """Hello"""] , foo_float=[0.1, 0.2, 0.3] ) , )
snake_case_ = parser.parse_args("""--foo_int 1 --bar_int 2 3 --foo_str a b c --foo_float 0.1 0.7""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo_int=[1] , bar_int=[2, 3] , foo_str=["""a""", """b""", """c"""] , foo_float=[0.1, 0.7] ) )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--bar""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--baz""" , default=UpperCAmelCase_ , type=UpperCAmelCase_ )
expected.add_argument("""--ces""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
expected.add_argument("""--des""" , nargs="""+""" , default=[] , type=UpperCAmelCase_ )
snake_case_ = [OptionalExample]
if is_python_no_less_than_3_10:
dataclass_types.append(UpperCAmelCase_ )
for dataclass_type in dataclass_types:
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_args([] )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=UpperCAmelCase_ , bar=UpperCAmelCase_ , baz=UpperCAmelCase_ , ces=[] , des=[] ) )
snake_case_ = parser.parse_args("""--foo 12 --bar 3.14 --baz 42 --ces a b c --des 1 2 3""".split() )
self.assertEqual(UpperCAmelCase_ , Namespace(foo=12 , bar=3.14 , baz="""42""" , ces=["""a""", """b""", """c"""] , des=[1, 2, 3] ) )
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--required_list""" , nargs="""+""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument("""--required_str""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = argparse.ArgumentParser()
expected.add_argument("""--foo""" , type=UpperCAmelCase_ , required=UpperCAmelCase_ )
expected.add_argument(
"""--required_enum""" , type=make_choice_type_function(["""titi""", """toto"""] ) , choices=["""titi""", """toto"""] , required=UpperCAmelCase_ , )
expected.add_argument("""--opt""" , type=UpperCAmelCase_ , default=UpperCAmelCase_ )
expected.add_argument("""--baz""" , default="""toto""" , type=UpperCAmelCase_ , help="""help message""" )
expected.add_argument("""--foo_str""" , nargs="""+""" , default=["""Hallo""", """Bonjour""", """Hello"""] , type=UpperCAmelCase_ )
self.argparsersEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Tuple:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
snake_case_ = parser.parse_dict(UpperCAmelCase_ )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
"""extra""": 42,
}
self.assertRaises(UpperCAmelCase_ , parser.parse_dict , UpperCAmelCase_ , allow_extra_keys=UpperCAmelCase_ )
def lowerCAmelCase ( self : Any ) ->Dict:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_json""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.json""" , """w+""" ) as f:
json.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.json""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Optional[int] ) ->List[str]:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
snake_case_ = {
"""foo""": 12,
"""bar""": 3.14,
"""baz""": """42""",
"""flag""": True,
}
with tempfile.TemporaryDirectory() as tmp_dir:
snake_case_ = os.path.join(UpperCAmelCase_ , """temp_yaml""" )
os.mkdir(UpperCAmelCase_ )
with open(temp_local_path + """.yaml""" , """w+""" ) as f:
yaml.dump(UpperCAmelCase_ , UpperCAmelCase_ )
snake_case_ = parser.parse_yaml_file(Path(temp_local_path + """.yaml""" ) )[0]
snake_case_ = BasicExample(**UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict ) ->Any:
"""simple docstring"""
snake_case_ = HfArgumentParser(UpperCAmelCase_ )
self.assertIsNotNone(UpperCAmelCase_ )
| 347 | 0 |
"""simple docstring"""
import baseaa
def _snake_case ( lowercase__ : str ) -> bytes:
'''simple docstring'''
return baseaa.aaaencode(string.encode("""utf-8""" ) )
def _snake_case ( lowercase__ : bytes ) -> str:
'''simple docstring'''
return baseaa.aaadecode(lowercase__ ).decode("""utf-8""" )
if __name__ == "__main__":
import doctest
doctest.testmod()
| 84 |
"""simple docstring"""
import logging
import os
from typing import Dict, List, Optional, Union
import torch
import torch.nn as nn
from accelerate.utils.imports import (
is_abit_bnb_available,
is_abit_bnb_available,
is_bnb_available,
)
from ..big_modeling import dispatch_model, init_empty_weights
from .dataclasses import BnbQuantizationConfig
from .modeling import (
find_tied_parameters,
get_balanced_memory,
infer_auto_device_map,
load_checkpoint_in_model,
offload_weight,
set_module_tensor_to_device,
)
if is_bnb_available():
import bitsandbytes as bnb
from copy import deepcopy
__SCREAMING_SNAKE_CASE : Any = logging.getLogger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = None , _SCREAMING_SNAKE_CASE = False , ) -> Optional[Any]:
snake_case_ = bnb_quantization_config.load_in_abit
snake_case_ = bnb_quantization_config.load_in_abit
if load_in_abit and not is_abit_bnb_available():
raise ImportError(
"""You have a version of `bitsandbytes` that is not compatible with 8bit quantization,"""
""" make sure you have the latest version of `bitsandbytes` installed.""" )
if load_in_abit and not is_abit_bnb_available():
raise ValueError(
"""You have a version of `bitsandbytes` that is not compatible with 4bit quantization,"""
"""make sure you have the latest version of `bitsandbytes` installed.""" )
snake_case_ = []
# custom device map
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) and len(device_map.keys() ) > 1:
snake_case_ = [key for key, value in device_map.items() if value in ["""disk""", """cpu"""]]
# We keep some modules such as the lm_head in their original dtype for numerical stability reasons
if bnb_quantization_config.skip_modules is None:
snake_case_ = get_keys_to_not_convert(_SCREAMING_SNAKE_CASE )
# add cpu modules to skip modules only for 4-bit modules
if load_in_abit:
bnb_quantization_config.skip_modules.extend(_SCREAMING_SNAKE_CASE )
snake_case_ = bnb_quantization_config.skip_modules
# We add the modules we want to keep in full precision
if bnb_quantization_config.keep_in_fpaa_modules is None:
snake_case_ = []
snake_case_ = bnb_quantization_config.keep_in_fpaa_modules
modules_to_not_convert.extend(_SCREAMING_SNAKE_CASE )
# compatibility with peft
snake_case_ = load_in_abit
snake_case_ = load_in_abit
snake_case_ = get_parameter_device(_SCREAMING_SNAKE_CASE )
if model_device.type != "meta":
# quantization of an already loaded model
logger.warning(
"""It is not recommended to quantize a loaded model. """
"""The model should be instantiated under the `init_empty_weights` context manager.""" )
snake_case_ = replace_with_bnb_layers(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
# convert param to the right dtype
snake_case_ = bnb_quantization_config.torch_dtype
for name, param in model.state_dict().items():
if any(module_to_keep_in_fpaa in name for module_to_keep_in_fpaa in keep_in_fpaa_modules ):
param.to(torch.floataa )
if param.dtype != torch.floataa:
snake_case_ = name.replace(""".weight""" , """""" ).replace(""".bias""" , """""" )
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if param is not None:
param.to(torch.floataa )
elif torch.is_floating_point(_SCREAMING_SNAKE_CASE ):
param.to(_SCREAMING_SNAKE_CASE )
if model_device.type == "cuda":
# move everything to cpu in the first place because we can't do quantization if the weights are already on cuda
model.cuda(torch.cuda.current_device() )
torch.cuda.empty_cache()
elif torch.cuda.is_available():
model.to(torch.cuda.current_device() )
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info(
f"""The model device type is {model_device.type}. However, cuda is needed for quantization."""
"""We move the model to cuda.""" )
return model
elif weights_location is None:
raise RuntimeError(
f"""`weights_location` needs to be the folder path containing the weights of the model, but we found {weights_location} """ )
else:
with init_empty_weights():
snake_case_ = replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , modules_to_not_convert=_SCREAMING_SNAKE_CASE )
snake_case_ = get_quantized_model_device_map(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , max_memory=_SCREAMING_SNAKE_CASE , no_split_module_classes=_SCREAMING_SNAKE_CASE , )
if offload_state_dict is None and device_map is not None and "disk" in device_map.values():
snake_case_ = True
snake_case_ = any(x in list(device_map.values() ) for x in ["""cpu""", """disk"""] )
load_checkpoint_in_model(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , dtype=bnb_quantization_config.torch_dtype , offload_folder=_SCREAMING_SNAKE_CASE , offload_state_dict=_SCREAMING_SNAKE_CASE , keep_in_fpaa_modules=bnb_quantization_config.keep_in_fpaa_modules , offload_abit_bnb=load_in_abit and offload , )
return dispatch_model(_SCREAMING_SNAKE_CASE , device_map=_SCREAMING_SNAKE_CASE , offload_dir=_SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if device_map is None:
if torch.cuda.is_available():
snake_case_ = {"""""": torch.cuda.current_device()}
else:
raise RuntimeError("""No GPU found. A GPU is needed for quantization.""" )
logger.info("""The device_map was not initialized.""" """Setting device_map to `{'':torch.cuda.current_device()}`.""" )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
if device_map not in ["auto", "balanced", "balanced_low_0", "sequential"]:
raise ValueError(
"""If passing a string for `device_map`, please choose 'auto', 'balanced', 'balanced_low_0' or """
"""'sequential'.""" )
snake_case_ = {}
special_dtypes.update(
{
name: bnb_quantization_config.torch_dtype
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.skip_modules )
} )
special_dtypes.update(
{
name: torch.floataa
for name, _ in model.named_parameters()
if any(m in name for m in bnb_quantization_config.keep_in_fpaa_modules )
} )
snake_case_ = {}
snake_case_ = special_dtypes
snake_case_ = no_split_module_classes
snake_case_ = bnb_quantization_config.target_dtype
# get max_memory for each device.
if device_map != "sequential":
snake_case_ = get_balanced_memory(
_SCREAMING_SNAKE_CASE , low_zero=(device_map == """balanced_low_0""") , max_memory=_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE , )
snake_case_ = max_memory
snake_case_ = infer_auto_device_map(_SCREAMING_SNAKE_CASE , **_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
# check if don't have any quantized module on the cpu
snake_case_ = bnb_quantization_config.skip_modules + bnb_quantization_config.keep_in_fpaa_modules
snake_case_ = {
key: device_map[key] for key in device_map.keys() if key not in modules_not_to_convert
}
for device in ["cpu", "disk"]:
if device in device_map_without_some_modules.values():
if bnb_quantization_config.load_in_abit:
raise ValueError(
"""
Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit
the quantized model. If you want to dispatch the model on the CPU or the disk while keeping
these modules in `torch_dtype`, you need to pass a custom `device_map` to
`load_and_quantize_model`. Check
https://huggingface.co/docs/accelerate/main/en/usage_guides/quantization#offload-modules-to-cpu-and-disk
for more details.
""" )
else:
logger.info(
"""Some modules are are offloaded to the CPU or the disk. Note that these modules will be converted to 8-bit""" )
del device_map_without_some_modules
return device_map
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None ) -> Tuple:
if modules_to_not_convert is None:
snake_case_ = []
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if not has_been_replaced:
logger.warning(
"""You are loading your model in 8bit or 4bit but no linear modules were found in your model."""
""" this can happen for some architectures such as gpt2 that uses Conv1D instead of Linear layers."""
""" Please double check your model architecture, or submit an issue on github if you think this is"""
""" a bug.""" )
return model
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None , _SCREAMING_SNAKE_CASE=None , ) -> List[Any]:
snake_case_ = False
for name, module in model.named_children():
if current_key_name is None:
snake_case_ = []
current_key_name.append(_SCREAMING_SNAKE_CASE )
if isinstance(_SCREAMING_SNAKE_CASE , nn.Linear ) and name not in modules_to_not_convert:
# Check if the current key is not in the `modules_to_not_convert`
snake_case_ = """.""".join(_SCREAMING_SNAKE_CASE )
snake_case_ = True
for key in modules_to_not_convert:
if (
(key in current_key_name_str) and (key + "." in current_key_name_str)
) or key == current_key_name_str:
snake_case_ = False
break
if proceed:
# Load bnb module with empty weight and replace ``nn.Linear` module
if bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.LinearabitLt(
module.in_features , module.out_features , module.bias is not None , has_fpaa_weights=_SCREAMING_SNAKE_CASE , threshold=bnb_quantization_config.llm_inta_threshold , )
elif bnb_quantization_config.load_in_abit:
snake_case_ = bnb.nn.Linearabit(
module.in_features , module.out_features , module.bias is not None , bnb_quantization_config.bnb_abit_compute_dtype , compress_statistics=bnb_quantization_config.bnb_abit_use_double_quant , quant_type=bnb_quantization_config.bnb_abit_quant_type , )
else:
raise ValueError("""load_in_8bit and load_in_4bit can't be both False""" )
snake_case_ = module.weight.data
if module.bias is not None:
snake_case_ = module.bias.data
bnb_module.requires_grad_(_SCREAMING_SNAKE_CASE )
setattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = True
if len(list(module.children() ) ) > 0:
snake_case_ , snake_case_ = _replace_with_bnb_layers(
_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
snake_case_ = has_been_replaced | _has_been_replaced
# Remove the last key for recursion
current_key_name.pop(-1 )
return model, has_been_replaced
def _a ( _SCREAMING_SNAKE_CASE ) -> Any:
# Create a copy of the model
with init_empty_weights():
snake_case_ = deepcopy(_SCREAMING_SNAKE_CASE ) # this has 0 cost since it is done inside `init_empty_weights` context manager`
snake_case_ = find_tied_parameters(_SCREAMING_SNAKE_CASE )
# For compatibility with Accelerate < 0.18
if isinstance(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ):
snake_case_ = sum(list(tied_params.values() ) , [] ) + list(tied_params.keys() )
else:
snake_case_ = sum(_SCREAMING_SNAKE_CASE , [] )
snake_case_ = len(_SCREAMING_SNAKE_CASE ) > 0
# Check if it is a base model
snake_case_ = False
if hasattr(_SCREAMING_SNAKE_CASE , """base_model_prefix""" ):
snake_case_ = not hasattr(_SCREAMING_SNAKE_CASE , model.base_model_prefix )
# Ignore this for base models (BertModel, GPT2Model, etc.)
if (not has_tied_params) and is_base_model:
return []
# otherwise they have an attached head
snake_case_ = list(model.named_children() )
snake_case_ = [list_modules[-1][0]]
# add last module together with tied weights
snake_case_ = set(_SCREAMING_SNAKE_CASE ) - set(_SCREAMING_SNAKE_CASE )
snake_case_ = list(set(_SCREAMING_SNAKE_CASE ) ) + list(_SCREAMING_SNAKE_CASE )
# remove ".weight" from the keys
snake_case_ = [""".weight""", """.bias"""]
snake_case_ = []
for name in list_untouched:
for name_to_remove in names_to_remove:
if name_to_remove in name:
snake_case_ = name.replace(_SCREAMING_SNAKE_CASE , """""" )
filtered_module_names.append(_SCREAMING_SNAKE_CASE )
return filtered_module_names
def _a ( _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for m in model.modules():
if isinstance(_SCREAMING_SNAKE_CASE , bnb.nn.Linearabit ):
return True
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[int]:
return next(parameter.parameters() ).device
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
# if it is not quantized, we quantize and offload the quantized weights and the SCB stats
if fpaa_statistics is None:
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , 0 , dtype=_SCREAMING_SNAKE_CASE , value=_SCREAMING_SNAKE_CASE )
snake_case_ = param_name
snake_case_ = model
if "." in tensor_name:
snake_case_ = tensor_name.split(""".""" )
for split in splits[:-1]:
snake_case_ = getattr(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if new_module is None:
raise ValueError(f"""{module} has no attribute {split}.""" )
snake_case_ = new_module
snake_case_ = splits[-1]
# offload weights
snake_case_ = False
offload_weight(module._parameters[tensor_name] , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
if hasattr(module._parameters[tensor_name] , """SCB""" ):
offload_weight(
module._parameters[tensor_name].SCB , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE , )
else:
offload_weight(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
offload_weight(_SCREAMING_SNAKE_CASE , param_name.replace("""weight""" , """SCB""" ) , _SCREAMING_SNAKE_CASE , index=_SCREAMING_SNAKE_CASE )
set_module_tensor_to_device(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , """meta""" , dtype=_SCREAMING_SNAKE_CASE , value=torch.empty(*param.size() ) )
| 347 | 0 |
'''simple docstring'''
_SCREAMING_SNAKE_CASE : Any = 0 # The first color of the flag.
_SCREAMING_SNAKE_CASE : Dict = 1 # The second color of the flag.
_SCREAMING_SNAKE_CASE : Tuple = 2 # The third color of the flag.
_SCREAMING_SNAKE_CASE : Union[str, Any] = (red, white, blue)
def UpperCamelCase_( snake_case : list ):
'''simple docstring'''
if not sequence:
return []
if len(snake_case ) == 1:
return list(snake_case )
snake_case_ = 0
snake_case_ = len(snake_case ) - 1
snake_case_ = 0
while mid <= high:
if sequence[mid] == colors[0]:
snake_case_ , snake_case_ = sequence[mid], sequence[low]
low += 1
mid += 1
elif sequence[mid] == colors[1]:
mid += 1
elif sequence[mid] == colors[2]:
snake_case_ , snake_case_ = sequence[high], sequence[mid]
high -= 1
else:
snake_case_ = f'The elements inside the sequence must contains only {colors} values'
raise ValueError(snake_case )
return sequence
if __name__ == "__main__":
import doctest
doctest.testmod()
_SCREAMING_SNAKE_CASE : Optional[int] = input("Enter numbers separated by commas:\n").strip()
_SCREAMING_SNAKE_CASE : Tuple = [int(item.strip()) for item in user_input.split(",")]
print(F"{dutch_national_flag_sort(unsorted)}")
| 85 |
"""simple docstring"""
from collections import OrderedDict
from typing import Mapping
from packaging import version
from ...configuration_utils import PretrainedConfig
from ...onnx import OnnxConfig
from ...utils import logging
__SCREAMING_SNAKE_CASE : str = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : Tuple = {
'microsoft/beit-base-patch16-224-pt22k': (
'https://huggingface.co/microsoft/beit-base-patch16-224-pt22k/resolve/main/config.json'
),
# See all BEiT models at https://huggingface.co/models?filter=beit
}
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = """beit"""
def __init__( self : List[str] , UpperCAmelCase_ : List[Any]=8_192 , UpperCAmelCase_ : Dict=768 , UpperCAmelCase_ : int=12 , UpperCAmelCase_ : Tuple=12 , UpperCAmelCase_ : List[Any]=3_072 , UpperCAmelCase_ : Tuple="gelu" , UpperCAmelCase_ : Dict=0.0 , UpperCAmelCase_ : List[str]=0.0 , UpperCAmelCase_ : Any=0.02 , UpperCAmelCase_ : Optional[Any]=1E-12 , UpperCAmelCase_ : int=224 , UpperCAmelCase_ : Tuple=16 , UpperCAmelCase_ : List[str]=3 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[str]=False , UpperCAmelCase_ : Tuple=False , UpperCAmelCase_ : int=False , UpperCAmelCase_ : List[Any]=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Any=True , UpperCAmelCase_ : Dict=[3, 5, 7, 11] , UpperCAmelCase_ : Tuple=[1, 2, 3, 6] , UpperCAmelCase_ : Optional[int]=True , UpperCAmelCase_ : List[Any]=0.4 , UpperCAmelCase_ : Optional[Any]=256 , UpperCAmelCase_ : Optional[Any]=1 , UpperCAmelCase_ : int=False , UpperCAmelCase_ : Tuple=255 , **UpperCAmelCase_ : List[str] , ) ->Optional[Any]:
"""simple docstring"""
super().__init__(**UpperCAmelCase_ )
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = initializer_range
snake_case_ = layer_norm_eps
snake_case_ = image_size
snake_case_ = patch_size
snake_case_ = num_channels
snake_case_ = use_mask_token
snake_case_ = use_absolute_position_embeddings
snake_case_ = use_relative_position_bias
snake_case_ = use_shared_relative_position_bias
snake_case_ = layer_scale_init_value
snake_case_ = drop_path_rate
snake_case_ = use_mean_pooling
# decode head attributes (semantic segmentation)
snake_case_ = out_indices
snake_case_ = pool_scales
# auxiliary head attributes (semantic segmentation)
snake_case_ = use_auxiliary_head
snake_case_ = auxiliary_loss_weight
snake_case_ = auxiliary_channels
snake_case_ = auxiliary_num_convs
snake_case_ = auxiliary_concat_input
snake_case_ = semantic_loss_ignore_index
class __A (snake_case__):
'''simple docstring'''
__lowercase: List[Any] = version.parse("""1.11""")
@property
def lowerCAmelCase ( self : Dict ) ->Mapping[str, Mapping[int, str]]:
"""simple docstring"""
return OrderedDict(
[
("""pixel_values""", {0: """batch""", 1: """num_channels""", 2: """height""", 3: """width"""}),
] )
@property
def lowerCAmelCase ( self : Any ) ->float:
"""simple docstring"""
return 1E-4
| 347 | 0 |
"""simple docstring"""
from __future__ import annotations
def __lowerCAmelCase (_UpperCamelCase ):
__lowerCAmelCase : List[Any] = 2
__lowerCAmelCase : List[Any] = []
while i * i <= n:
if n % i:
i += 1
else:
n //= i
factors.append(_UpperCamelCase )
if n > 1:
factors.append(_UpperCamelCase )
return factors
if __name__ == "__main__":
import doctest
doctest.testmod() | 86 |
"""simple docstring"""
import os
import re
import warnings
from shutil import copyfile
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
import sentencepiece as spm
from ...tokenization_utils import PreTrainedTokenizer
if TYPE_CHECKING:
from ...tokenization_utils_base import TextInput
from ...utils import logging
__SCREAMING_SNAKE_CASE : Union[str, Any] = logging.get_logger(__name__)
__SCREAMING_SNAKE_CASE : List[Any] = {'vocab_file': 'spiece.model'}
__SCREAMING_SNAKE_CASE : int = {
'vocab_file': {
't5-small': 'https://huggingface.co/t5-small/resolve/main/spiece.model',
't5-base': 'https://huggingface.co/t5-base/resolve/main/spiece.model',
't5-large': 'https://huggingface.co/t5-large/resolve/main/spiece.model',
't5-3b': 'https://huggingface.co/t5-3b/resolve/main/spiece.model',
't5-11b': 'https://huggingface.co/t5-11b/resolve/main/spiece.model',
}
}
# TODO(PVP) - this should be removed in Transformers v5
__SCREAMING_SNAKE_CASE : Dict = {
't5-small': 512,
't5-base': 512,
't5-large': 512,
't5-3b': 512,
't5-11b': 512,
}
__SCREAMING_SNAKE_CASE : Optional[int] = '▁'
class __A (snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = VOCAB_FILES_NAMES
__lowercase: Any = PRETRAINED_VOCAB_FILES_MAP
__lowercase: Any = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
__lowercase: List[str] = ["""input_ids""", """attention_mask"""]
def __init__( self : Optional[int] , UpperCAmelCase_ : Tuple , UpperCAmelCase_ : Optional[Any]="</s>" , UpperCAmelCase_ : Optional[Any]="<unk>" , UpperCAmelCase_ : Any="<pad>" , UpperCAmelCase_ : Tuple=100 , UpperCAmelCase_ : Optional[Any]=None , UpperCAmelCase_ : Optional[Dict[str, Any]] = None , UpperCAmelCase_ : Optional[int]=True , **UpperCAmelCase_ : Dict , ) ->None:
"""simple docstring"""
if extra_ids > 0 and additional_special_tokens is None:
snake_case_ = [F"""<extra_id_{i}>""" for i in range(UpperCAmelCase_ )]
elif extra_ids > 0 and additional_special_tokens is not None:
# Check that we have the right number of extra_id special tokens
snake_case_ = len(set(filter(lambda UpperCAmelCase_ : bool("""extra_id""" in str(UpperCAmelCase_ ) ) , UpperCAmelCase_ ) ) )
if extra_tokens != extra_ids:
raise ValueError(
F"""Both extra_ids ({extra_ids}) and additional_special_tokens ({additional_special_tokens}) are"""
""" provided to T5Tokenizer. In this case the additional_special_tokens must include the extra_ids"""
""" tokens""" )
if legacy:
logger.warning_once(
F"""You are using the legacy behaviour of the {self.__class__}. This means that tokens that come after special tokens will not be properly handled. We recommend you to"""
""" read the related pull request available at https://github.com/huggingface/transformers/pull/24565""" )
snake_case_ = legacy
snake_case_ = {} if sp_model_kwargs is None else sp_model_kwargs
super().__init__(
eos_token=UpperCAmelCase_ , unk_token=UpperCAmelCase_ , pad_token=UpperCAmelCase_ , extra_ids=UpperCAmelCase_ , additional_special_tokens=UpperCAmelCase_ , sp_model_kwargs=self.sp_model_kwargs , legacy=UpperCAmelCase_ , **UpperCAmelCase_ , )
snake_case_ = vocab_file
snake_case_ = extra_ids
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(UpperCAmelCase_ )
@staticmethod
def lowerCAmelCase ( UpperCAmelCase_ : str , UpperCAmelCase_ : Dict , UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
if pretrained_model_name_or_path in TaTokenizer.max_model_input_sizes:
snake_case_ = TaTokenizer.max_model_input_sizes[pretrained_model_name_or_path]
if init_max_model_length is not None and init_max_model_length != max_model_length:
return init_max_model_length
elif init_max_model_length is None:
warnings.warn(
"""This tokenizer was incorrectly instantiated with a model max length of"""
F""" {deprecated_max_model_length} which will be corrected in Transformers v5.\nFor now, this"""
""" behavior is kept to avoid breaking backwards compatibility when padding/encoding with"""
""" `truncation is True`.\n- Be aware that you SHOULD NOT rely on"""
F""" {pretrained_model_name_or_path} automatically truncating your input to"""
F""" {deprecated_max_model_length} when padding/encoding.\n- If you want to encode/pad to sequences"""
F""" longer than {deprecated_max_model_length} you can either instantiate this tokenizer with"""
""" `model_max_length` or pass `max_length` when encoding/padding.\n- To avoid this warning, please"""
""" instantiate this tokenizer with `model_max_length` set to your preferred value.""" , UpperCAmelCase_ , )
return max_model_length
@property
def lowerCAmelCase ( self : Optional[Any] ) ->Optional[Any]:
"""simple docstring"""
return self.sp_model.get_piece_size() + self._extra_ids
def lowerCAmelCase ( self : Any ) ->Optional[int]:
"""simple docstring"""
snake_case_ = {self.convert_ids_to_tokens(UpperCAmelCase_ ): i for i in range(self.vocab_size )}
vocab.update(self.added_tokens_encoder )
return vocab
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None , UpperCAmelCase_ : bool = False ) ->List[int]:
"""simple docstring"""
if already_has_special_tokens:
return super().get_special_tokens_mask(
token_ids_a=UpperCAmelCase_ , token_ids_a=UpperCAmelCase_ , already_has_special_tokens=UpperCAmelCase_ )
# normal case: some special tokens
if token_ids_a is None:
return ([0] * len(UpperCAmelCase_ )) + [1]
return ([0] * len(UpperCAmelCase_ )) + [1] + ([0] * len(UpperCAmelCase_ )) + [1]
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
return list(
set(filter(lambda UpperCAmelCase_ : bool(re.search(R"""<extra_id_\d+>""" , UpperCAmelCase_ ) ) is not None , self.additional_special_tokens ) ) )
def lowerCAmelCase ( self : Dict ) ->str:
"""simple docstring"""
return [self._convert_token_to_id(UpperCAmelCase_ ) for token in self.get_sentinel_tokens()]
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : List[int] ) ->List[int]:
"""simple docstring"""
if len(UpperCAmelCase_ ) > 0 and token_ids[-1] == self.eos_token_id:
warnings.warn(
F"""This sequence already has {self.eos_token}. In future versions this behavior may lead to duplicated"""
""" eos tokens being added.""" )
return token_ids
else:
return token_ids + [self.eos_token_id]
def lowerCAmelCase ( self : str , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = [self.eos_token_id]
if token_ids_a is None:
return len(token_ids_a + eos ) * [0]
return len(token_ids_a + eos + token_ids_a + eos ) * [0]
def lowerCAmelCase ( self : Optional[int] , UpperCAmelCase_ : List[int] , UpperCAmelCase_ : Optional[List[int]] = None ) ->List[int]:
"""simple docstring"""
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
if token_ids_a is None:
return token_ids_a
else:
snake_case_ = self._add_eos_if_not_present(UpperCAmelCase_ )
return token_ids_a + token_ids_a
def __getstate__( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = self.__dict__.copy()
snake_case_ = None
return state
def __setstate__( self : Optional[Any] , UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
snake_case_ = d
# for backward compatibility
if not hasattr(self , """sp_model_kwargs""" ):
snake_case_ = {}
snake_case_ = spm.SentencePieceProcessor(**self.sp_model_kwargs )
self.sp_model.Load(self.vocab_file )
def lowerCAmelCase ( self : int , UpperCAmelCase_ : "TextInput" , **UpperCAmelCase_ : Tuple ) ->List[str]:
"""simple docstring"""
if not self.legacy:
snake_case_ = SPIECE_UNDERLINE + text.replace(UpperCAmelCase_ , """ """ )
return super().tokenize(UpperCAmelCase_ , **UpperCAmelCase_ )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Any ) ->Tuple:
"""simple docstring"""
if not self.legacy:
snake_case_ = text.startswith(UpperCAmelCase_ )
if is_first:
snake_case_ = text[1:]
snake_case_ = self.sp_model.encode(UpperCAmelCase_ , out_type=UpperCAmelCase_ )
if not self.legacy and not is_first and not text.startswith(""" """ ) and tokens[0].startswith(UpperCAmelCase_ ):
snake_case_ = ([tokens[0][1:]] if len(tokens[0] ) > 1 else []) + tokens[1:]
return tokens
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
if token.startswith("""<extra_id_""" ):
snake_case_ = re.match(R"""<extra_id_(\d+)>""" , UpperCAmelCase_ )
snake_case_ = int(match.group(1 ) )
return self.vocab_size - num - 1
return self.sp_model.piece_to_id(UpperCAmelCase_ )
def lowerCAmelCase ( self : List[str] , UpperCAmelCase_ : Optional[Any] ) ->List[Any]:
"""simple docstring"""
if index < self.sp_model.get_piece_size():
snake_case_ = self.sp_model.IdToPiece(UpperCAmelCase_ )
else:
snake_case_ = F"""<extra_id_{self.vocab_size - 1 - index}>"""
return token
def lowerCAmelCase ( self : List[Any] , UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = []
snake_case_ = """"""
snake_case_ = False
for token in tokens:
# make sure that special tokens are not decoded using sentencepiece model
if token in self.all_special_tokens:
if not prev_is_special:
out_string += " "
out_string += self.sp_model.decode(UpperCAmelCase_ ) + token
snake_case_ = True
snake_case_ = []
else:
current_sub_tokens.append(UpperCAmelCase_ )
snake_case_ = False
out_string += self.sp_model.decode(UpperCAmelCase_ )
return out_string.strip()
def lowerCAmelCase ( self : str , UpperCAmelCase_ : str , UpperCAmelCase_ : Optional[str] = None ) ->Tuple[str]:
"""simple docstring"""
if not os.path.isdir(UpperCAmelCase_ ):
logger.error(F"""Vocabulary path ({save_directory}) should be a directory""" )
return
snake_case_ = os.path.join(
UpperCAmelCase_ , (filename_prefix + """-""" if filename_prefix else """""") + VOCAB_FILES_NAMES["""vocab_file"""] )
if os.path.abspath(self.vocab_file ) != os.path.abspath(UpperCAmelCase_ ) and os.path.isfile(self.vocab_file ):
copyfile(self.vocab_file , UpperCAmelCase_ )
elif not os.path.isfile(self.vocab_file ):
with open(UpperCAmelCase_ , """wb""" ) as fi:
snake_case_ = self.sp_model.serialized_model_proto()
fi.write(UpperCAmelCase_ )
return (out_vocab_file,)
| 347 | 0 |
from __future__ import annotations
from math import gcd
def lowercase_ ( _lowerCamelCase : int , _lowerCamelCase : int = 2 , _lowerCamelCase : int = 1 , _lowerCamelCase : int = 3 , ):
# A value less than 2 can cause an infinite loop in the algorithm.
if num < 2:
raise ValueError("The input value cannot be less than 2")
# Because of the relationship between ``f(f(x))`` and ``f(x)``, this
# algorithm struggles to find factors that are divisible by two.
# As a workaround, we specifically check for two and even inputs.
# See: https://math.stackexchange.com/a/2856214/165820
if num > 2 and num % 2 == 0:
return 2
# Pollard's Rho algorithm requires a function that returns pseudorandom
# values between 0 <= X < ``num``. It doesn't need to be random in the
# sense that the output value is cryptographically secure or difficult
# to calculate, it only needs to be random in the sense that all output
# values should be equally likely to appear.
# For this reason, Pollard suggested using ``f(x) = (x**2 - 1) % num``
# However, the success of Pollard's algorithm isn't guaranteed and is
# determined in part by the initial seed and the chosen random function.
# To make retries easier, we will instead use ``f(x) = (x**2 + C) % num``
# where ``C`` is a value that we can modify between each attempt.
def rand_fn(_lowerCamelCase : int , _lowerCamelCase : int , _lowerCamelCase : int) -> int:
return (pow(_lowerCamelCase , 2) + step) % modulus
for _ in range(_lowerCamelCase):
# These track the position within the cycle detection logic.
lowercase__ : Optional[int] = seed
lowercase__ : int = seed
while True:
# At each iteration, the tortoise moves one step and the hare moves two.
lowercase__ : List[Any] = rand_fn(_lowerCamelCase , _lowerCamelCase , _lowerCamelCase)
lowercase__ : List[str] = rand_fn(_lowerCamelCase , _lowerCamelCase , _lowerCamelCase)
lowercase__ : int = rand_fn(_lowerCamelCase , _lowerCamelCase , _lowerCamelCase)
# At some point both the tortoise and the hare will enter a cycle whose
# length ``p`` is a divisor of ``num``. Once in that cycle, at some point
# the tortoise and hare will end up on the same value modulo ``p``.
# We can detect when this happens because the position difference between
# the tortoise and the hare will share a common divisor with ``num``.
lowercase__ : Any = gcd(hare - tortoise , _lowerCamelCase)
if divisor == 1:
# No common divisor yet, just keep searching.
continue
else:
# We found a common divisor!
if divisor == num:
# Unfortunately, the divisor is ``num`` itself and is useless.
break
else:
# The divisor is a nontrivial factor of ``num``!
return divisor
# If we made it here, then this attempt failed.
# We need to pick a new starting seed for the tortoise and hare
# in addition to a new step value for the random function.
# To keep this example implementation deterministic, the
# new values will be generated based on currently available
# values instead of using something like ``random.randint``.
# We can use the hare's position as the new seed.
# This is actually what Richard Brent's the "optimized" variant does.
lowercase__ : Dict = hare
# The new step value for the random function can just be incremented.
# At first the results will be similar to what the old function would
# have produced, but the value will quickly diverge after a bit.
step += 1
# We haven't found a divisor within the requested number of attempts.
# We were unlucky or ``num`` itself is actually prime.
return None
if __name__ == "__main__":
import argparse
UpperCamelCase = argparse.ArgumentParser()
parser.add_argument(
'''num''',
type=int,
help='''The value to find a divisor of''',
)
parser.add_argument(
'''--attempts''',
type=int,
default=3,
help='''The number of attempts before giving up''',
)
UpperCamelCase = parser.parse_args()
UpperCamelCase = pollard_rho(args.num, attempts=args.attempts)
if divisor is None:
print(f"{args.num} is probably prime")
else:
UpperCamelCase = args.num // divisor
print(f"{args.num} = {divisor} * {quotient}")
| 87 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE = 1_000_000 ) -> int:
snake_case_ = [i - 1 for i in range(limit + 1 )]
for i in range(2 , limit + 1 ):
if phi[i] == i - 1:
for j in range(2 * i , limit + 1 , _SCREAMING_SNAKE_CASE ):
phi[j] -= phi[j] // i
return sum(phi[2 : limit + 1] )
if __name__ == "__main__":
print(solution())
| 347 | 0 |
import json
from typing import TYPE_CHECKING, List, Optional, Tuple
from tokenizers import pre_tokenizers
from ...tokenization_utils_base import BatchEncoding
from ...tokenization_utils_fast import PreTrainedTokenizerFast
from ...utils import logging
if TYPE_CHECKING:
from transformers.pipelines.conversational import Conversation
__lowerCAmelCase : List[Any] = logging.get_logger(__name__)
__lowerCAmelCase : List[str] = {'tokenizer_file': 'tokenizer.json'}
__lowerCAmelCase : Optional[Any] = {
'tokenizer_file': {
'bigscience/tokenizer': 'https://huggingface.co/bigscience/tokenizer/blob/main/tokenizer.json',
'bigscience/bloom-560m': 'https://huggingface.co/bigscience/bloom-560m/blob/main/tokenizer.json',
'bigscience/bloom-1b1': 'https://huggingface.co/bigscience/bloom-1b1/blob/main/tokenizer.json',
'bigscience/bloom-1b7': 'https://huggingface.co/bigscience/bloom-1b7/blob/main/tokenizer.json',
'bigscience/bloom-3b': 'https://huggingface.co/bigscience/bloom-3b/blob/main/tokenizer.json',
'bigscience/bloom-7b1': 'https://huggingface.co/bigscience/bloom-7b1/blob/main/tokenizer.json',
'bigscience/bloom': 'https://huggingface.co/bigscience/bloom/blob/main/tokenizer.json',
},
}
class UpperCAmelCase_ ( _A ):
'''simple docstring'''
a__ = VOCAB_FILES_NAMES
a__ = PRETRAINED_VOCAB_FILES_MAP
a__ = ["""input_ids""", """attention_mask"""]
a__ = None
def __init__( self : Optional[Any] , UpperCamelCase__ : List[str]=None , UpperCamelCase__ : Optional[int]=None , UpperCamelCase__ : str=None , UpperCamelCase__ : Optional[Any]="<unk>" , UpperCamelCase__ : Optional[Any]="<s>" , UpperCamelCase__ : List[str]="</s>" , UpperCamelCase__ : List[str]="<pad>" , UpperCamelCase__ : List[str]=False , UpperCamelCase__ : List[Any]=False , **UpperCamelCase__ : Dict , ) -> int:
"""simple docstring"""
super().__init__(
UpperCamelCase__ , UpperCamelCase__ , tokenizer_file=UpperCamelCase__ , unk_token=UpperCamelCase__ , bos_token=UpperCamelCase__ , eos_token=UpperCamelCase__ , pad_token=UpperCamelCase__ , add_prefix_space=UpperCamelCase__ , clean_up_tokenization_spaces=UpperCamelCase__ , **UpperCamelCase__ , )
__magic_name__ = json.loads(self.backend_tokenizer.pre_tokenizer.__getstate__() )
if pre_tok_state.get("""add_prefix_space""" , UpperCamelCase__ ) != add_prefix_space:
__magic_name__ = getattr(UpperCamelCase__ , pre_tok_state.pop("""type""" ) )
__magic_name__ = add_prefix_space
__magic_name__ = pre_tok_class(**UpperCamelCase__ )
__magic_name__ = add_prefix_space
def _lowercase ( self : Optional[Any] , *UpperCamelCase__ : List[Any] , **UpperCamelCase__ : str ) -> BatchEncoding:
"""simple docstring"""
__magic_name__ = kwargs.get("""is_split_into_words""" , UpperCamelCase__ )
if not (self.add_prefix_space or not is_split_into_words):
raise Exception(
F'''You need to instantiate {self.__class__.__name__} with add_prefix_space=True to use it with'''
""" pretokenized inputs.""" )
return super()._batch_encode_plus(*UpperCamelCase__ , **UpperCamelCase__ )
def _lowercase ( self : int , *UpperCamelCase__ : str , **UpperCamelCase__ : Union[str, Any] ) -> BatchEncoding:
"""simple docstring"""
__magic_name__ = kwargs.get("""is_split_into_words""" , UpperCamelCase__ )
if not (self.add_prefix_space or not is_split_into_words):
raise Exception(
F'''You need to instantiate {self.__class__.__name__} with add_prefix_space=True to use it with'''
""" pretokenized inputs.""" )
return super()._encode_plus(*UpperCamelCase__ , **UpperCamelCase__ )
def _lowercase ( self : str , UpperCamelCase__ : str , UpperCamelCase__ : Optional[str] = None ) -> Tuple[str]:
"""simple docstring"""
__magic_name__ = self._tokenizer.model.save(UpperCamelCase__ , name=UpperCamelCase__ )
return tuple(UpperCamelCase__ )
def _lowercase ( self : Union[str, Any] , UpperCamelCase__ : "Conversation" ) -> List[int]:
"""simple docstring"""
__magic_name__ = []
for is_user, text in conversation.iter_texts():
input_ids.extend(self.encode(UpperCamelCase__ , add_special_tokens=UpperCamelCase__ ) + [self.eos_token_id] )
if len(UpperCamelCase__ ) > self.model_max_length:
__magic_name__ = input_ids[-self.model_max_length :]
return input_ids
| 88 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
print(f"""Vertex\tShortest Distance from vertex {src}""" )
for i, d in enumerate(_SCREAMING_SNAKE_CASE ):
print(f"""{i}\t\t{d}""" )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Union[str, Any]:
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
return True
return False
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> list[float]:
snake_case_ = [float("""inf""" )] * vertex_count
snake_case_ = 0.0
for _ in range(vertex_count - 1 ):
for j in range(_SCREAMING_SNAKE_CASE ):
snake_case_ , snake_case_ , snake_case_ = (graph[j][k] for k in ["""src""", """dst""", """weight"""])
if distance[u] != float("""inf""" ) and distance[u] + w < distance[v]:
snake_case_ = distance[u] + w
snake_case_ = check_negative_cycle(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
if negative_cycle_exists:
raise Exception("""Negative cycle found""" )
return distance
if __name__ == "__main__":
import doctest
doctest.testmod()
__SCREAMING_SNAKE_CASE : int = int(input('Enter number of vertices: ').strip())
__SCREAMING_SNAKE_CASE : Dict = int(input('Enter number of edges: ').strip())
__SCREAMING_SNAKE_CASE : list[dict[str, int]] = [{} for _ in range(E)]
for i in range(E):
print('Edge ', i + 1)
__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE : List[str] = (
int(x)
for x in input('Enter source, destination, weight: ').strip().split(' ')
)
__SCREAMING_SNAKE_CASE : Union[str, Any] = {'src': src, 'dst': dest, 'weight': weight}
__SCREAMING_SNAKE_CASE : Union[str, Any] = int(input('\nEnter shortest path source:').strip())
__SCREAMING_SNAKE_CASE : str = bellman_ford(graph, V, E, source)
print_distance(shortest_distance, 0)
| 347 | 0 |
'''simple docstring'''
import inspect
import unittest
from transformers import ViTMSNConfig
from transformers.testing_utils import require_torch, require_vision, slow, torch_device
from transformers.utils import cached_property, is_torch_available, is_vision_available
from ...test_configuration_common import ConfigTester
from ...test_modeling_common import ModelTesterMixin, floats_tensor, ids_tensor
from ...test_pipeline_mixin import PipelineTesterMixin
if is_torch_available():
import torch
from torch import nn
from transformers import ViTMSNForImageClassification, ViTMSNModel
from transformers.models.vit_msn.modeling_vit_msn import VIT_MSN_PRETRAINED_MODEL_ARCHIVE_LIST
if is_vision_available():
from PIL import Image
from transformers import ViTImageProcessor
class __magic_name__ :
def __init__( self : Optional[int] ,_UpperCAmelCase : Optional[int] ,_UpperCAmelCase : Union[str, Any]=13 ,_UpperCAmelCase : Optional[int]=30 ,_UpperCAmelCase : Union[str, Any]=2 ,_UpperCAmelCase : List[str]=3 ,_UpperCAmelCase : Any=True ,_UpperCAmelCase : Tuple=True ,_UpperCAmelCase : Tuple=32 ,_UpperCAmelCase : Union[str, Any]=5 ,_UpperCAmelCase : str=4 ,_UpperCAmelCase : int=37 ,_UpperCAmelCase : int="gelu" ,_UpperCAmelCase : List[str]=0.1 ,_UpperCAmelCase : Optional[Any]=0.1 ,_UpperCAmelCase : List[Any]=10 ,_UpperCAmelCase : str=0.02 ,_UpperCAmelCase : List[str]=None ,):
_a : int = parent
_a : str = batch_size
_a : Tuple = image_size
_a : str = patch_size
_a : List[str] = num_channels
_a : Union[str, Any] = is_training
_a : Union[str, Any] = use_labels
_a : List[Any] = hidden_size
_a : Tuple = num_hidden_layers
_a : List[str] = num_attention_heads
_a : List[Any] = intermediate_size
_a : Union[str, Any] = hidden_act
_a : Optional[Any] = hidden_dropout_prob
_a : Dict = attention_probs_dropout_prob
_a : Optional[Any] = type_sequence_label_size
_a : List[str] = initializer_range
_a : List[Any] = scope
# in ViT MSN, the seq length equals the number of patches + 1 (we add 1 for the [CLS] token)
_a : int = (image_size // patch_size) ** 2
_a : List[str] = num_patches + 1
def __lowercase ( self : str ):
_a : Tuple = floats_tensor([self.batch_size, self.num_channels, self.image_size, self.image_size] )
_a : Any = None
if self.use_labels:
_a : str = ids_tensor([self.batch_size] ,self.type_sequence_label_size )
_a : List[str] = self.get_config()
return config, pixel_values, labels
def __lowercase ( self : List[str] ):
return ViTMSNConfig(
image_size=self.image_size ,patch_size=self.patch_size ,num_channels=self.num_channels ,hidden_size=self.hidden_size ,num_hidden_layers=self.num_hidden_layers ,num_attention_heads=self.num_attention_heads ,intermediate_size=self.intermediate_size ,hidden_act=self.hidden_act ,hidden_dropout_prob=self.hidden_dropout_prob ,attention_probs_dropout_prob=self.attention_probs_dropout_prob ,initializer_range=self.initializer_range ,)
def __lowercase ( self : Tuple ,_UpperCAmelCase : Any ,_UpperCAmelCase : Optional[Any] ,_UpperCAmelCase : Any ):
_a : Optional[int] = ViTMSNModel(config=_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
_a : Tuple = model(_UpperCAmelCase )
self.parent.assertEqual(result.last_hidden_state.shape ,(self.batch_size, self.seq_length, self.hidden_size) )
def __lowercase ( self : Optional[Any] ,_UpperCAmelCase : Dict ,_UpperCAmelCase : Union[str, Any] ,_UpperCAmelCase : Any ):
_a : Optional[int] = self.type_sequence_label_size
_a : Any = ViTMSNForImageClassification(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
_a : Optional[Any] = model(_UpperCAmelCase ,labels=_UpperCAmelCase )
print('Pixel and labels shape: {pixel_values.shape}, {labels.shape}' )
print('Labels: {labels}' )
self.parent.assertEqual(result.logits.shape ,(self.batch_size, self.type_sequence_label_size) )
# test greyscale images
_a : Tuple = 1
_a : Optional[Any] = ViTMSNForImageClassification(_UpperCAmelCase )
model.to(_UpperCAmelCase )
model.eval()
_a : Optional[int] = floats_tensor([self.batch_size, 1, self.image_size, self.image_size] )
_a : Union[str, Any] = model(_UpperCAmelCase )
self.parent.assertEqual(result.logits.shape ,(self.batch_size, self.type_sequence_label_size) )
def __lowercase ( self : Dict ):
_a : int = self.prepare_config_and_inputs()
_a , _a , _a : List[str] = config_and_inputs
_a : Dict = {'pixel_values': pixel_values}
return config, inputs_dict
@require_torch
class __magic_name__ ( _UpperCamelCase , _UpperCamelCase , unittest.TestCase ):
lowerCAmelCase : Any = (ViTMSNModel, ViTMSNForImageClassification) if is_torch_available() else ()
lowerCAmelCase : Optional[int] = (
{'feature-extraction': ViTMSNModel, 'image-classification': ViTMSNForImageClassification}
if is_torch_available()
else {}
)
lowerCAmelCase : List[Any] = False
lowerCAmelCase : Optional[int] = False
lowerCAmelCase : Optional[int] = False
lowerCAmelCase : Optional[int] = False
def __lowercase ( self : Tuple ):
_a : List[Any] = ViTMSNModelTester(self )
_a : Dict = ConfigTester(self ,config_class=_UpperCAmelCase ,has_text_modality=_UpperCAmelCase ,hidden_size=37 )
def __lowercase ( self : Optional[Any] ):
self.config_tester.run_common_tests()
@unittest.skip(reason='ViTMSN does not use inputs_embeds' )
def __lowercase ( self : str ):
pass
def __lowercase ( self : List[str] ):
_a , _a : Any = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
_a : int = model_class(_UpperCAmelCase )
self.assertIsInstance(model.get_input_embeddings() ,(nn.Module) )
_a : int = model.get_output_embeddings()
self.assertTrue(x is None or isinstance(_UpperCAmelCase ,nn.Linear ) )
def __lowercase ( self : Optional[int] ):
_a , _a : Optional[int] = self.model_tester.prepare_config_and_inputs_for_common()
for model_class in self.all_model_classes:
_a : int = model_class(_UpperCAmelCase )
_a : Tuple = inspect.signature(model.forward )
# signature.parameters is an OrderedDict => so arg_names order is deterministic
_a : List[str] = [*signature.parameters.keys()]
_a : List[Any] = ['pixel_values']
self.assertListEqual(arg_names[:1] ,_UpperCAmelCase )
def __lowercase ( self : int ):
_a : List[str] = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_model(*_UpperCAmelCase )
def __lowercase ( self : Dict ):
_a : Dict = self.model_tester.prepare_config_and_inputs()
self.model_tester.create_and_check_for_image_classification(*_UpperCAmelCase )
@slow
def __lowercase ( self : List[Any] ):
for model_name in VIT_MSN_PRETRAINED_MODEL_ARCHIVE_LIST[:1]:
_a : List[str] = ViTMSNModel.from_pretrained(_UpperCAmelCase )
self.assertIsNotNone(_UpperCAmelCase )
def __lowerCamelCase ( ) -> Tuple:
_a : Tuple = Image.open('./tests/fixtures/tests_samples/COCO/000000039769.png' )
return image
@require_torch
@require_vision
class __magic_name__ ( unittest.TestCase ):
@cached_property
def __lowercase ( self : Optional[Any] ):
return ViTImageProcessor.from_pretrained('facebook/vit-msn-small' ) if is_vision_available() else None
@slow
def __lowercase ( self : Tuple ):
torch.manual_seed(2 )
_a : List[str] = ViTMSNForImageClassification.from_pretrained('facebook/vit-msn-small' ).to(_UpperCAmelCase )
_a : List[str] = self.default_image_processor
_a : Dict = prepare_img()
_a : Union[str, Any] = image_processor(images=_UpperCAmelCase ,return_tensors='pt' ).to(_UpperCAmelCase )
# forward pass
with torch.no_grad():
_a : List[str] = model(**_UpperCAmelCase )
# verify the logits
_a : Optional[int] = torch.Size((1, 1000) )
self.assertEqual(outputs.logits.shape ,_UpperCAmelCase )
_a : Any = torch.tensor([-0.08_03, -0.44_54, -0.23_75] ).to(_UpperCAmelCase )
self.assertTrue(torch.allclose(outputs.logits[0, :3] ,_UpperCAmelCase ,atol=1E-4 ) )
| 89 |
"""simple docstring"""
import argparse
import logging
import os
import re
import tensorflow as tf
from transformers import (
AutoConfig,
AutoTokenizer,
DataCollatorForLanguageModeling,
PushToHubCallback,
TFAutoModelForMaskedLM,
create_optimizer,
)
__SCREAMING_SNAKE_CASE : List[str] = logging.getLogger(__name__)
__SCREAMING_SNAKE_CASE : str = tf.data.AUTOTUNE
def _a ( ) -> List[str]:
snake_case_ = argparse.ArgumentParser(description="""Train a masked language model on TPU.""" )
parser.add_argument(
"""--pretrained_model_config""" , type=_SCREAMING_SNAKE_CASE , default="""roberta-base""" , help="""The model config to use. Note that we don't copy the model's weights, only the config!""" , )
parser.add_argument(
"""--tokenizer""" , type=_SCREAMING_SNAKE_CASE , default="""unigram-tokenizer-wikitext""" , help="""The name of the tokenizer to load. We use the pretrained tokenizer to initialize the model's vocab size.""" , )
parser.add_argument(
"""--per_replica_batch_size""" , type=_SCREAMING_SNAKE_CASE , default=8 , help="""Batch size per TPU core.""" , )
parser.add_argument(
"""--no_tpu""" , action="""store_true""" , help="""If set, run on CPU and don't try to initialize a TPU. Useful for debugging on non-TPU instances.""" , )
parser.add_argument(
"""--tpu_name""" , type=_SCREAMING_SNAKE_CASE , help="""Name of TPU resource to initialize. Should be blank on Colab, and 'local' on TPU VMs.""" , default="""local""" , )
parser.add_argument(
"""--tpu_zone""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud zone that TPU resource is located in. Only used for non-Colab TPU nodes.""" , )
parser.add_argument(
"""--gcp_project""" , type=_SCREAMING_SNAKE_CASE , help="""Google cloud project name. Only used for non-Colab TPU nodes.""" )
parser.add_argument(
"""--bfloat16""" , action="""store_true""" , help="""Use mixed-precision bfloat16 for training. This is the recommended lower-precision format for TPU.""" , )
parser.add_argument(
"""--train_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to training dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--shuffle_buffer_size""" , type=_SCREAMING_SNAKE_CASE , default=2**18 , help="""Size of the shuffle buffer (in samples)""" , )
parser.add_argument(
"""--eval_dataset""" , type=_SCREAMING_SNAKE_CASE , help="""Path to evaluation dataset to load. If the path begins with `gs://`"""
""" then the dataset will be loaded from a Google Cloud Storage bucket.""" , )
parser.add_argument(
"""--num_epochs""" , type=_SCREAMING_SNAKE_CASE , default=1 , help="""Number of epochs to train for.""" , )
parser.add_argument(
"""--learning_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-4 , help="""Learning rate to use for training.""" , )
parser.add_argument(
"""--weight_decay_rate""" , type=_SCREAMING_SNAKE_CASE , default=1E-3 , help="""Weight decay rate to use for training.""" , )
parser.add_argument(
"""--max_length""" , type=_SCREAMING_SNAKE_CASE , default=512 , help="""Maximum length of tokenized sequences. Should match the setting used in prepare_tfrecord_shards.py""" , )
parser.add_argument(
"""--mlm_probability""" , type=_SCREAMING_SNAKE_CASE , default=0.15 , help="""Fraction of tokens to mask during training.""" , )
parser.add_argument("""--output_dir""" , type=_SCREAMING_SNAKE_CASE , required=_SCREAMING_SNAKE_CASE , help="""Path to save model checkpoints to.""" )
parser.add_argument("""--hub_model_id""" , type=_SCREAMING_SNAKE_CASE , help="""Model ID to upload to on the Hugging Face Hub.""" )
snake_case_ = parser.parse_args()
return args
def _a ( _SCREAMING_SNAKE_CASE ) -> Optional[Any]:
try:
if args.tpu_name:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver(
args.tpu_name , zone=args.tpu_zone , project=args.gcp_project )
else:
snake_case_ = tf.distribute.cluster_resolver.TPUClusterResolver()
except ValueError:
raise RuntimeError(
"""Couldn't connect to TPU! Most likely you need to specify --tpu_name, --tpu_zone, or """
"""--gcp_project. When running on a TPU VM, use --tpu_name local.""" )
tf.config.experimental_connect_to_cluster(_SCREAMING_SNAKE_CASE )
tf.tpu.experimental.initialize_tpu_system(_SCREAMING_SNAKE_CASE )
return tpu
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = 0
for file in file_list:
snake_case_ = file.split("""/""" )[-1]
snake_case_ = re.search(r"""-\d+-(\d+)\.tfrecord""" , _SCREAMING_SNAKE_CASE ).group(1 )
snake_case_ = int(_SCREAMING_SNAKE_CASE )
num_samples += sample_count
return num_samples
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=None ) -> Union[str, Any]:
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.data.Dataset.from_tensor_slices(_SCREAMING_SNAKE_CASE )
if shuffle:
snake_case_ = dataset.shuffle(len(_SCREAMING_SNAKE_CASE ) )
snake_case_ = tf.data.TFRecordDataset(_SCREAMING_SNAKE_CASE , num_parallel_reads=_SCREAMING_SNAKE_CASE )
# TF can't infer the total sample count because it doesn't read all the records yet, so we assert it here
snake_case_ = dataset.apply(tf.data.experimental.assert_cardinality(_SCREAMING_SNAKE_CASE ) )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
if shuffle:
assert shuffle_buffer_size is not None
snake_case_ = dataset.shuffle(args.shuffle_buffer_size )
snake_case_ = dataset.batch(_SCREAMING_SNAKE_CASE , drop_remainder=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.map(_SCREAMING_SNAKE_CASE , num_parallel_calls=_SCREAMING_SNAKE_CASE )
snake_case_ = dataset.prefetch(_SCREAMING_SNAKE_CASE )
return dataset
def _a ( _SCREAMING_SNAKE_CASE ) -> List[Any]:
if not args.no_tpu:
snake_case_ = initialize_tpu(_SCREAMING_SNAKE_CASE )
snake_case_ = tf.distribute.TPUStrategy(_SCREAMING_SNAKE_CASE )
else:
snake_case_ = tf.distribute.OneDeviceStrategy(device="""/gpu:0""" )
if args.bfloataa:
tf.keras.mixed_precision.set_global_policy("""mixed_bfloat16""" )
snake_case_ = AutoTokenizer.from_pretrained(args.tokenizer )
snake_case_ = AutoConfig.from_pretrained(args.pretrained_model_config )
snake_case_ = tokenizer.vocab_size
snake_case_ = tf.io.gfile.glob(os.path.join(args.train_dataset , """*.tfrecord""" ) )
if not training_records:
raise ValueError(f"""No .tfrecord files found in {args.train_dataset}.""" )
snake_case_ = tf.io.gfile.glob(os.path.join(args.eval_dataset , """*.tfrecord""" ) )
if not eval_records:
raise ValueError(f"""No .tfrecord files found in {args.eval_dataset}.""" )
snake_case_ = count_samples(_SCREAMING_SNAKE_CASE )
snake_case_ = num_train_samples // (args.per_replica_batch_size * strategy.num_replicas_in_sync)
snake_case_ = steps_per_epoch * args.num_epochs
with strategy.scope():
snake_case_ = TFAutoModelForMaskedLM.from_config(_SCREAMING_SNAKE_CASE )
model(model.dummy_inputs ) # Pass some dummy inputs through the model to ensure all the weights are built
snake_case_ , snake_case_ = create_optimizer(
num_train_steps=_SCREAMING_SNAKE_CASE , num_warmup_steps=total_train_steps // 20 , init_lr=args.learning_rate , weight_decay_rate=args.weight_decay_rate , )
# Transformers models compute the right loss for their task by default when labels are passed, and will
# use this for training unless you specify your own loss function in compile().
model.compile(optimizer=_SCREAMING_SNAKE_CASE , metrics=["""accuracy"""] )
def decode_fn(_SCREAMING_SNAKE_CASE ):
snake_case_ = {
"""input_ids""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
"""attention_mask""": tf.io.FixedLenFeature(dtype=tf.intaa , shape=(args.max_length,) ),
}
return tf.io.parse_single_example(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# Many of the data collators in Transformers are TF-compilable when return_tensors == "tf", so we can
# use their methods in our data pipeline.
snake_case_ = DataCollatorForLanguageModeling(
tokenizer=_SCREAMING_SNAKE_CASE , mlm_probability=args.mlm_probability , mlm=_SCREAMING_SNAKE_CASE , return_tensors="""tf""" )
def mask_with_collator(_SCREAMING_SNAKE_CASE ):
# TF really needs an isin() function
snake_case_ = (
~tf.cast(batch["""attention_mask"""] , tf.bool )
| (batch["""input_ids"""] == tokenizer.cls_token_id)
| (batch["""input_ids"""] == tokenizer.sep_token_id)
)
snake_case_ , snake_case_ = data_collator.tf_mask_tokens(
batch["""input_ids"""] , vocab_size=len(_SCREAMING_SNAKE_CASE ) , mask_token_id=tokenizer.mask_token_id , special_tokens_mask=_SCREAMING_SNAKE_CASE , )
return batch
snake_case_ = args.per_replica_batch_size * strategy.num_replicas_in_sync
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , shuffle_buffer_size=args.shuffle_buffer_size , )
snake_case_ = prepare_dataset(
_SCREAMING_SNAKE_CASE , decode_fn=_SCREAMING_SNAKE_CASE , mask_fn=_SCREAMING_SNAKE_CASE , batch_size=_SCREAMING_SNAKE_CASE , shuffle=_SCREAMING_SNAKE_CASE , )
snake_case_ = []
if args.hub_model_id:
callbacks.append(
PushToHubCallback(output_dir=args.output_dir , hub_model_id=args.hub_model_id , tokenizer=_SCREAMING_SNAKE_CASE ) )
model.fit(
_SCREAMING_SNAKE_CASE , validation_data=_SCREAMING_SNAKE_CASE , epochs=args.num_epochs , callbacks=_SCREAMING_SNAKE_CASE , )
model.save_pretrained(args.output_dir )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : Union[str, Any] = parse_args()
main(args)
| 347 | 0 |
import argparse
import requests
import torch
# pip3 install salesforce-lavis
# I'm actually installing a slightly modified version: pip3 install git+https://github.com/nielsrogge/LAVIS.git@fix_lavis_float32 (there's also the fix_lavis branch)
# also note: to convert Vicuna checkpoints, we had to include /home/niels/python_projects/checkpoints/FastChat/vicuna-7b in lavis/configs/models/blip2/blip2_instruct_vicuna7b.yaml
# same for Vicuna-13b
from lavis.models import load_model_and_preprocess
from PIL import Image
from transformers import (
AutoTokenizer,
BlipImageProcessor,
InstructBlipConfig,
InstructBlipForConditionalGeneration,
InstructBlipProcessor,
InstructBlipQFormerConfig,
InstructBlipVisionConfig,
LlamaConfig,
LlamaTokenizerFast,
TaConfig,
TaTokenizerFast,
)
from transformers.utils.constants import OPENAI_CLIP_MEAN, OPENAI_CLIP_STD
def lowerCamelCase_ ( ) -> List[str]:
"""simple docstring"""
__lowerCamelCase = 'https://raw.githubusercontent.com/salesforce/LAVIS/main/docs/_static/Confusing-Pictures.jpg'
__lowerCamelCase = Image.open(requests.get(UpperCamelCase__ , stream=UpperCamelCase__ ).raw ).convert('RGB' )
return image
def lowerCamelCase_ ( UpperCamelCase__ : Any ) -> Optional[Any]:
"""simple docstring"""
__lowerCamelCase = []
# fmt: off
# vision encoder
rename_keys.append(('visual_encoder.cls_token', 'vision_model.embeddings.class_embedding') )
rename_keys.append(('visual_encoder.pos_embed', 'vision_model.embeddings.position_embedding') )
rename_keys.append(('visual_encoder.patch_embed.proj.weight', 'vision_model.embeddings.patch_embedding.weight') )
rename_keys.append(('visual_encoder.patch_embed.proj.bias', 'vision_model.embeddings.patch_embedding.bias') )
rename_keys.append(('ln_vision.weight', 'vision_model.post_layernorm.weight') )
rename_keys.append(('ln_vision.bias', 'vision_model.post_layernorm.bias') )
for i in range(config.vision_config.num_hidden_layers ):
rename_keys.append((F"""visual_encoder.blocks.{i}.norm1.weight""", F"""vision_model.encoder.layers.{i}.layer_norm1.weight""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.norm1.bias""", F"""vision_model.encoder.layers.{i}.layer_norm1.bias""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.norm2.weight""", F"""vision_model.encoder.layers.{i}.layer_norm2.weight""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.norm2.bias""", F"""vision_model.encoder.layers.{i}.layer_norm2.bias""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.attn.qkv.weight""", F"""vision_model.encoder.layers.{i}.self_attn.qkv.weight""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.attn.proj.weight""", F"""vision_model.encoder.layers.{i}.self_attn.projection.weight""",) )
rename_keys.append((F"""visual_encoder.blocks.{i}.attn.proj.bias""", F"""vision_model.encoder.layers.{i}.self_attn.projection.bias""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.mlp.fc1.weight""", F"""vision_model.encoder.layers.{i}.mlp.fc1.weight""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.mlp.fc1.bias""", F"""vision_model.encoder.layers.{i}.mlp.fc1.bias""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.mlp.fc2.weight""", F"""vision_model.encoder.layers.{i}.mlp.fc2.weight""") )
rename_keys.append((F"""visual_encoder.blocks.{i}.mlp.fc2.bias""", F"""vision_model.encoder.layers.{i}.mlp.fc2.bias""") )
# QFormer
rename_keys.append(('Qformer.bert.embeddings.LayerNorm.weight', 'qformer.embeddings.layernorm.weight') )
rename_keys.append(('Qformer.bert.embeddings.LayerNorm.bias', 'qformer.embeddings.layernorm.bias') )
# fmt: on
return rename_keys
def lowerCamelCase_ ( UpperCamelCase__ : List[Any] , UpperCamelCase__ : str , UpperCamelCase__ : Optional[int] ) -> Optional[int]:
"""simple docstring"""
__lowerCamelCase = dct.pop(UpperCamelCase__ )
__lowerCamelCase = val
def lowerCamelCase_ ( UpperCamelCase__ : Tuple , UpperCamelCase__ : int ) -> Dict:
"""simple docstring"""
for i in range(config.vision_config.num_hidden_layers ):
# read in original q and v biases
__lowerCamelCase = state_dict.pop(F"""visual_encoder.blocks.{i}.attn.q_bias""" )
__lowerCamelCase = state_dict.pop(F"""visual_encoder.blocks.{i}.attn.v_bias""" )
# next, set bias in the state dict
__lowerCamelCase = torch.cat((q_bias, torch.zeros_like(UpperCamelCase__ , requires_grad=UpperCamelCase__ ), v_bias) )
__lowerCamelCase = qkv_bias
def lowerCamelCase_ ( UpperCamelCase__ : Dict ) -> Optional[Any]:
"""simple docstring"""
__lowerCamelCase = 364 if 'coco' in model_name else 224
__lowerCamelCase = InstructBlipVisionConfig(image_size=UpperCamelCase__ ).to_dict()
# make sure the models have proper bos_token_id and eos_token_id set (important for generation)
# seems like flan-T5 models don't have bos_token_id properly set?
if "t5-xl" in model_name:
__lowerCamelCase = TaConfig.from_pretrained('google/flan-t5-xl' , dense_act_fn='gelu' , bos_token_id=1 ).to_dict()
elif "t5-xxl" in model_name:
__lowerCamelCase = TaConfig.from_pretrained('google/flan-t5-xxl' , dense_act_fn='gelu' , bos_token_id=1 ).to_dict()
elif "vicuna-7b" in model_name:
__lowerCamelCase = LlamaConfig.from_pretrained('decapoda-research/llama-7b-hf' , vocab_size=3_2001 ).to_dict()
elif "vicuna-13b" in model_name:
__lowerCamelCase = LlamaConfig.from_pretrained('decapoda-research/llama-13b-hf' , vocab_size=3_2001 ).to_dict()
else:
raise ValueError('Model name not supported' )
# the authors add one special "[DEC]" token to the vocab of Q-Former, hence vocab size = 30522 + 1
__lowerCamelCase = InstructBlipQFormerConfig(vocab_size=3_0523 ).to_dict()
__lowerCamelCase = InstructBlipConfig(vision_config=UpperCamelCase__ , text_config=UpperCamelCase__ , qformer_config=UpperCamelCase__ )
return config, image_size
@torch.no_grad()
def lowerCamelCase_ ( UpperCamelCase__ : Dict , UpperCamelCase__ : Optional[int]=None , UpperCamelCase__ : Union[str, Any]=False ) -> Optional[Any]:
"""simple docstring"""
__lowerCamelCase = AutoTokenizer.from_pretrained('bert-base-uncased' , truncation_side='left' )
qformer_tokenizer.add_special_tokens({'bos_token': '[DEC]'} )
if "t5" in model_name:
__lowerCamelCase = TaTokenizerFast.from_pretrained('google/flan-t5-xl' , truncation_side='left' )
elif "vicuna" in model_name:
# the following was used in the original implementation:
# tokenizer = LlamaTokenizer.from_pretrained("huggyllama/llama-7b", use_fast=False, truncation_side="left")
# tokenizer.add_special_tokens({"pad_token": "[PAD]"})
# tokenizer.add_special_tokens({"bos_token": "</s>"})
# tokenizer.add_special_tokens({"eos_token": "</s>"})
# tokenizer.add_special_tokens({"unk_token": "</s>"})
__lowerCamelCase = LlamaTokenizerFast.from_pretrained(
'huggyllama/llama-7b' , truncation_side='left' , bos_token='</s>' , unk_token='</s>' )
tokenizer.add_special_tokens({'pad_token': '[PAD]'} )
__lowerCamelCase , __lowerCamelCase = get_blipa_config(UpperCamelCase__ )
__lowerCamelCase = InstructBlipForConditionalGeneration(UpperCamelCase__ ).eval()
__lowerCamelCase = {
'instructblip-vicuna-7b': ('blip2_vicuna_instruct', 'vicuna7b'),
'instructblip-vicuna-13b': ('blip2_vicuna_instruct', 'vicuna13b'),
'instructblip-flan-t5-xl': ('blip2_t5_instruct', 'flant5xl'),
'instructblip-flan-t5-xxl': ('blip2_t5_instruct', 'flant5xxl'),
}
__lowerCamelCase , __lowerCamelCase = model_name_to_original[model_name]
# load original model
print('Loading original model...' )
__lowerCamelCase = 'cuda:1' if torch.cuda.is_available() else 'cpu'
__lowerCamelCase = 'cuda:2' if torch.cuda.is_available() else 'cpu'
__lowerCamelCase , __lowerCamelCase , __lowerCamelCase = load_model_and_preprocess(
name=UpperCamelCase__ , model_type=UpperCamelCase__ , is_eval=UpperCamelCase__ , device=UpperCamelCase__ )
original_model.eval()
print('Done!' )
# update state dict keys
__lowerCamelCase = original_model.state_dict()
__lowerCamelCase = create_rename_keys(UpperCamelCase__ )
for src, dest in rename_keys:
rename_key(UpperCamelCase__ , UpperCamelCase__ , UpperCamelCase__ )
# some keys can be renamed efficiently
for key, val in state_dict.copy().items():
__lowerCamelCase = state_dict.pop(UpperCamelCase__ )
if key.startswith('Qformer.bert' ):
__lowerCamelCase = key.replace('Qformer.bert' , 'qformer' )
if "attention.self" in key:
__lowerCamelCase = key.replace('self' , 'attention' )
if "llm_proj" in key:
__lowerCamelCase = key.replace('llm_proj' , 'language_projection' )
if "t5_proj" in key:
__lowerCamelCase = key.replace('t5_proj' , 'language_projection' )
if key.startswith('llm_model' ):
__lowerCamelCase = key.replace('llm_model' , 'language_model' )
if key.startswith('t5' ):
__lowerCamelCase = key.replace('t5' , 'language' )
__lowerCamelCase = val
# read in qv biases
read_in_q_v_bias(UpperCamelCase__ , UpperCamelCase__ )
# note: weights get loaded in torch.float32 by default
hf_model.load_state_dict(UpperCamelCase__ , strict=UpperCamelCase__ )
__lowerCamelCase = load_demo_image()
__lowerCamelCase = 'What is unusual about this image?'
# create processor
__lowerCamelCase = BlipImageProcessor(
size={'height': image_size, 'width': image_size} , image_mean=UpperCamelCase__ , image_std=UpperCamelCase__ )
__lowerCamelCase = InstructBlipProcessor(
image_processor=UpperCamelCase__ , tokenizer=UpperCamelCase__ , qformer_tokenizer=UpperCamelCase__ , )
__lowerCamelCase = processor(images=UpperCamelCase__ , text=UpperCamelCase__ , return_tensors='pt' ).to(UpperCamelCase__ )
# make sure processor creates exact same pixel values
__lowerCamelCase = vis_processors['eval'](UpperCamelCase__ ).unsqueeze(0 ).to(UpperCamelCase__ )
__lowerCamelCase = inputs.pixel_values
assert torch.allclose(original_pixel_values.to(pixel_values.device ) , UpperCamelCase__ )
original_model.to(UpperCamelCase__ )
hf_model.to(UpperCamelCase__ )
with torch.no_grad():
if "vicuna" in model_name:
__lowerCamelCase = original_model({'image': original_pixel_values, 'text_input': [prompt]} ).logits
__lowerCamelCase = hf_model(**UpperCamelCase__ ).logits
else:
__lowerCamelCase = original_model(
{'image': original_pixel_values, 'text_input': [prompt], 'text_output': ['\n']} ).logits
__lowerCamelCase = tokenizer('\n' , return_tensors='pt' ).input_ids.to(UpperCamelCase__ )
__lowerCamelCase = label_input_ids.masked_fill(label_input_ids == tokenizer.pad_token_id , -100 )
__lowerCamelCase = hf_model(**UpperCamelCase__ , labels=UpperCamelCase__ ).logits
print('First values of original logits:' , original_logits[0, :3, :3] )
print('First values of HF logits:' , logits[0, :3, :3] )
# assert values
assert original_logits.shape == logits.shape
__lowerCamelCase = 1E-4 if 'vicuna' in model_name else 1E-5
assert torch.allclose(original_logits.to(logits.device ) , UpperCamelCase__ , atol=UpperCamelCase__ )
print('Looks ok!' )
print('Generating with original model...' )
__lowerCamelCase = original_model.generate({'image': original_pixel_values, 'prompt': prompt} , num_beams=5 )
# important: we need to cast the weights of the HF model to the appropriate type
print('Generating with HF model...' )
__lowerCamelCase = hf_model.generate(
**UpperCamelCase__ , do_sample=UpperCamelCase__ , num_beams=5 , max_length=256 , min_length=1 , top_p=0.9 , repetition_penalty=1.5 , length_penalty=1.0 , temperature=1 , )
if "vicuna" in model_name:
# convert output id 0 to 2 (eos_token_id)
# TODO add this in the generate method?
__lowerCamelCase = 2
print('Original generation:' , UpperCamelCase__ )
__lowerCamelCase = processor.batch_decode(UpperCamelCase__ , skip_special_tokens=UpperCamelCase__ )
__lowerCamelCase = [text.strip() for text in output_text]
print('HF generation:' , UpperCamelCase__ )
if pytorch_dump_folder_path is not None:
processor.save_pretrained(UpperCamelCase__ )
hf_model.save_pretrained(UpperCamelCase__ )
if push_to_hub:
processor.push_to_hub(F"""Salesforce/{model_name}""" )
hf_model.push_to_hub(F"""Salesforce/{model_name}""" )
if __name__ == "__main__":
__A = argparse.ArgumentParser()
__A = [
"instructblip-vicuna-7b",
"instructblip-vicuna-13b",
"instructblip-flan-t5-xl",
"instructblip-flan-t5-xxl",
]
parser.add_argument(
"--model_name",
default="instructblip-flan-t5-xl",
choices=choices,
type=str,
help="Path to hf config.json of model to convert",
)
parser.add_argument("--pytorch_dump_folder_path", default=None, type=str, help="Path to the output PyTorch model.")
parser.add_argument(
"--push_to_hub",
action="store_true",
help="Whether to push the model and processor to the hub after converting",
)
__A = parser.parse_args()
convert_blipa_checkpoint(args.model_name, args.pytorch_dump_folder_path, args.push_to_hub)
| 90 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> float:
if density <= 0:
raise ValueError("""Impossible fluid density""" )
if bulk_modulus <= 0:
raise ValueError("""Impossible bulk modulus""" )
return (bulk_modulus / density) ** 0.5
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
"""simple docstring"""
from typing import List, Union
from ..utils import (
add_end_docstrings,
is_tf_available,
is_torch_available,
is_vision_available,
logging,
requires_backends,
)
from .base import PIPELINE_INIT_ARGS, Pipeline
if is_vision_available():
from PIL import Image
from ..image_utils import load_image
if is_tf_available():
from ..models.auto.modeling_tf_auto import TF_MODEL_FOR_VISION_2_SEQ_MAPPING
if is_torch_available():
import torch
from ..models.auto.modeling_auto import MODEL_FOR_VISION_2_SEQ_MAPPING
UpperCAmelCase_ : Any = logging.get_logger(__name__)
@add_end_docstrings(UpperCAmelCase__ )
class lowerCAmelCase__ ( UpperCAmelCase__ ):
'''simple docstring'''
def __init__( self : Tuple , *lowercase_ : str , **lowercase_ : List[Any]):
'''simple docstring'''
super().__init__(*lowercase_ , **lowercase_)
requires_backends(self , '''vision''')
self.check_model_type(
TF_MODEL_FOR_VISION_2_SEQ_MAPPING if self.framework == '''tf''' else MODEL_FOR_VISION_2_SEQ_MAPPING)
def _SCREAMING_SNAKE_CASE ( self : Any , lowercase_ : str=None , lowercase_ : Dict=None , lowercase_ : Optional[int]=None):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : List[Any] = {}
SCREAMING_SNAKE_CASE_ : Tuple = {}
if prompt is not None:
SCREAMING_SNAKE_CASE_ : Tuple = prompt
if generate_kwargs is not None:
SCREAMING_SNAKE_CASE_ : int = generate_kwargs
if max_new_tokens is not None:
if "generate_kwargs" not in forward_kwargs:
SCREAMING_SNAKE_CASE_ : Union[str, Any] = {}
if "max_new_tokens" in forward_kwargs["generate_kwargs"]:
raise ValueError(
'''\'max_new_tokens\' is defined twice, once in \'generate_kwargs\' and once as a direct parameter,'''
''' please use only one''')
SCREAMING_SNAKE_CASE_ : Optional[Any] = max_new_tokens
return preprocess_params, forward_kwargs, {}
def __call__( self : Any , lowercase_ : Union[str, List[str], "Image.Image", List["Image.Image"]] , **lowercase_ : Any):
'''simple docstring'''
return super().__call__(lowercase_ , **lowercase_)
def _SCREAMING_SNAKE_CASE ( self : Dict , lowercase_ : Optional[int] , lowercase_ : List[str]=None):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : int = load_image(lowercase_)
if prompt is not None:
if not isinstance(lowercase_ , lowercase_):
raise ValueError(
F'Received an invalid text input, got - {type(lowercase_)} - but expected a single string. '
'''Note also that one single text can be provided for conditional image to text generation.''')
SCREAMING_SNAKE_CASE_ : Optional[int] = self.model.config.model_type
if model_type == "git":
SCREAMING_SNAKE_CASE_ : Dict = self.image_processor(images=lowercase_ , return_tensors=self.framework)
SCREAMING_SNAKE_CASE_ : Optional[Any] = self.tokenizer(text=lowercase_ , add_special_tokens=lowercase_).input_ids
SCREAMING_SNAKE_CASE_ : Optional[int] = [self.tokenizer.cls_token_id] + input_ids
SCREAMING_SNAKE_CASE_ : Tuple = torch.tensor(lowercase_).unsqueeze(0)
model_inputs.update({'''input_ids''': input_ids})
elif model_type == "pix2struct":
SCREAMING_SNAKE_CASE_ : str = self.image_processor(images=lowercase_ , header_text=lowercase_ , return_tensors=self.framework)
elif model_type != "vision-encoder-decoder":
# vision-encoder-decoder does not support conditional generation
SCREAMING_SNAKE_CASE_ : int = self.image_processor(images=lowercase_ , return_tensors=self.framework)
SCREAMING_SNAKE_CASE_ : int = self.tokenizer(lowercase_ , return_tensors=self.framework)
model_inputs.update(lowercase_)
else:
raise ValueError(F'Model type {model_type} does not support conditional text generation')
else:
SCREAMING_SNAKE_CASE_ : List[str] = self.image_processor(images=lowercase_ , return_tensors=self.framework)
if self.model.config.model_type == "git" and prompt is None:
SCREAMING_SNAKE_CASE_ : List[str] = None
return model_inputs
def _SCREAMING_SNAKE_CASE ( self : Any , lowercase_ : List[str] , lowercase_ : Optional[int]=None):
'''simple docstring'''
if (
"input_ids" in model_inputs
and isinstance(model_inputs['''input_ids'''] , lowercase_)
and all(x is None for x in model_inputs['''input_ids'''])
):
SCREAMING_SNAKE_CASE_ : List[str] = None
if generate_kwargs is None:
SCREAMING_SNAKE_CASE_ : Optional[int] = {}
# FIXME: We need to pop here due to a difference in how `generation.py` and `generation.tf_utils.py`
# parse inputs. In the Tensorflow version, `generate` raises an error if we don't use `input_ids` whereas
# the PyTorch version matches it with `self.model.main_input_name` or `self.model.encoder.main_input_name`
# in the `_prepare_model_inputs` method.
SCREAMING_SNAKE_CASE_ : int = model_inputs.pop(self.model.main_input_name)
SCREAMING_SNAKE_CASE_ : Optional[int] = self.model.generate(lowercase_ , **lowercase_ , **lowercase_)
return model_outputs
def _SCREAMING_SNAKE_CASE ( self : List[Any] , lowercase_ : Any):
'''simple docstring'''
SCREAMING_SNAKE_CASE_ : Union[str, Any] = []
for output_ids in model_outputs:
SCREAMING_SNAKE_CASE_ : Optional[Any] = {
'''generated_text''': self.tokenizer.decode(
lowercase_ , skip_special_tokens=lowercase_ , )
}
records.append(lowercase_)
return records
| 91 |
"""simple docstring"""
def _a ( _SCREAMING_SNAKE_CASE ) -> bool:
if num < 0:
return False
snake_case_ = num
snake_case_ = 0
while num > 0:
snake_case_ = rev_num * 10 + (num % 10)
num //= 10
return num_copy == rev_num
if __name__ == "__main__":
import doctest
doctest.testmod()
| 347 | 0 |
from ...utils import is_note_seq_available, is_transformers_available, is_torch_available
from ...utils import OptionalDependencyNotAvailable
try:
if not (is_transformers_available() and is_torch_available()):
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
from ...utils.dummy_torch_and_transformers_objects import * # noqa F403
else:
from .notes_encoder import SpectrogramNotesEncoder
from .continous_encoder import SpectrogramContEncoder
from .pipeline_spectrogram_diffusion import (
SpectrogramContEncoder,
SpectrogramDiffusionPipeline,
TaFilmDecoder,
)
try:
if not (is_transformers_available() and is_torch_available() and is_note_seq_available()):
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
from ...utils.dummy_transformers_and_torch_and_note_seq_objects import * # noqa F403
else:
from .midi_utils import MidiProcessor
| 92 |
"""simple docstring"""
import unittest
from transformers import SPIECE_UNDERLINE
from transformers.models.speechta import SpeechTaTokenizer
from transformers.testing_utils import get_tests_dir, require_sentencepiece, require_tokenizers, slow
from transformers.tokenization_utils import AddedToken
from ...test_tokenization_common import TokenizerTesterMixin
__SCREAMING_SNAKE_CASE : Tuple = get_tests_dir('fixtures/test_sentencepiece_bpe_char.model')
@require_sentencepiece
@require_tokenizers
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Tuple = SpeechTaTokenizer
__lowercase: int = False
__lowercase: List[str] = True
def lowerCAmelCase ( self : Any ) ->str:
"""simple docstring"""
super().setUp()
# We have a SentencePiece fixture for testing
snake_case_ = SpeechTaTokenizer(UpperCAmelCase_ )
snake_case_ = AddedToken("""<mask>""" , lstrip=UpperCAmelCase_ , rstrip=UpperCAmelCase_ )
snake_case_ = mask_token
tokenizer.add_special_tokens({"""mask_token""": mask_token} )
tokenizer.add_tokens(["""<ctc_blank>"""] )
tokenizer.save_pretrained(self.tmpdirname )
def lowerCAmelCase ( self : Optional[Any] , UpperCAmelCase_ : Optional[Any] ) ->Dict:
"""simple docstring"""
snake_case_ = """this is a test"""
snake_case_ = """this is a test"""
return input_text, output_text
def lowerCAmelCase ( self : str , UpperCAmelCase_ : int , UpperCAmelCase_ : Any=False , UpperCAmelCase_ : Tuple=20 , UpperCAmelCase_ : Dict=5 ) ->List[Any]:
"""simple docstring"""
snake_case_ , snake_case_ = self.get_input_output_texts(UpperCAmelCase_ )
snake_case_ = tokenizer.encode(UpperCAmelCase_ , add_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.decode(UpperCAmelCase_ , clean_up_tokenization_spaces=UpperCAmelCase_ )
return text, ids
def lowerCAmelCase ( self : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
snake_case_ = """<pad>"""
snake_case_ = 1
self.assertEqual(self.get_tokenizer()._convert_token_to_id(UpperCAmelCase_ ) , UpperCAmelCase_ )
self.assertEqual(self.get_tokenizer()._convert_id_to_token(UpperCAmelCase_ ) , UpperCAmelCase_ )
def lowerCAmelCase ( self : int ) ->str:
"""simple docstring"""
snake_case_ = list(self.get_tokenizer().get_vocab().keys() )
self.assertEqual(vocab_keys[0] , """<s>""" )
self.assertEqual(vocab_keys[1] , """<pad>""" )
self.assertEqual(vocab_keys[-4] , """œ""" )
self.assertEqual(vocab_keys[-2] , """<mask>""" )
self.assertEqual(vocab_keys[-1] , """<ctc_blank>""" )
self.assertEqual(len(UpperCAmelCase_ ) , 81 )
def lowerCAmelCase ( self : Optional[int] ) ->int:
"""simple docstring"""
self.assertEqual(self.get_tokenizer().vocab_size , 79 )
def lowerCAmelCase ( self : Optional[int] ) ->List[Any]:
"""simple docstring"""
snake_case_ = self.get_tokenizers(do_lower_case=UpperCAmelCase_ )
for tokenizer in tokenizers:
with self.subTest(F"""{tokenizer.__class__.__name__}""" ):
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
# We usually have added tokens from the start in tests because our vocab fixtures are
# smaller than the original vocabs - let's not assert this
# self.assertEqual(vocab_size, all_size)
snake_case_ = ["""aaaaa bbbbbb""", """cccccccccdddddddd"""]
snake_case_ = tokenizer.add_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode("""aaaaa bbbbbb low cccccccccdddddddd l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 4 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
snake_case_ = {"""eos_token""": """>>>>|||<||<<|<<""", """pad_token""": """<<<<<|||>|>>>>|>"""}
snake_case_ = tokenizer.add_special_tokens(UpperCAmelCase_ )
snake_case_ = tokenizer.vocab_size
snake_case_ = len(UpperCAmelCase_ )
self.assertNotEqual(UpperCAmelCase_ , 0 )
self.assertEqual(UpperCAmelCase_ , UpperCAmelCase_ )
self.assertEqual(UpperCAmelCase_ , len(UpperCAmelCase_ ) )
self.assertEqual(UpperCAmelCase_ , all_size_a + len(UpperCAmelCase_ ) )
snake_case_ = tokenizer.encode(
""">>>>|||<||<<|<< aaaaabbbbbb low cccccccccdddddddd <<<<<|||>|>>>>|> l""" , add_special_tokens=UpperCAmelCase_ )
self.assertGreaterEqual(len(UpperCAmelCase_ ) , 6 )
self.assertGreater(tokens[0] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[0] , tokens[1] )
self.assertGreater(tokens[-3] , tokenizer.vocab_size - 1 )
self.assertGreater(tokens[-3] , tokens[-4] )
self.assertEqual(tokens[0] , tokenizer.eos_token_id )
self.assertEqual(tokens[-3] , tokenizer.pad_token_id )
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->Optional[Any]:
"""simple docstring"""
pass
def lowerCAmelCase ( self : List[str] ) ->List[str]:
"""simple docstring"""
snake_case_ = self.get_tokenizer()
snake_case_ = tokenizer.tokenize("""This is a test""" )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [SPIECE_UNDERLINE, """T""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """a""", SPIECE_UNDERLINE, """t""", """e""", """s""", """t"""] )
# fmt: on
self.assertListEqual(
tokenizer.convert_tokens_to_ids(UpperCAmelCase_ ) , [4, 32, 11, 10, 12, 4, 10, 12, 4, 7, 4, 6, 5, 12, 6] , )
snake_case_ = tokenizer.tokenize("""I was born in 92000, and this is falsé.""" )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """92000""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
snake_case_ = tokenizer.convert_tokens_to_ids(UpperCAmelCase_ )
# fmt: off
self.assertListEqual(UpperCAmelCase_ , [4, 30, 4, 20, 7, 12, 4, 25, 8, 13, 9, 4, 10, 9, 4, 3, 23, 4, 7, 9, 14, 4, 6, 11, 10, 12, 4, 10, 12, 4, 19, 7, 15, 12, 73, 26] )
# fmt: on
snake_case_ = tokenizer.convert_ids_to_tokens(UpperCAmelCase_ )
self.assertListEqual(
UpperCAmelCase_ , [SPIECE_UNDERLINE, """I""", SPIECE_UNDERLINE, """w""", """a""", """s""", SPIECE_UNDERLINE, """b""", """o""", """r""", """n""", SPIECE_UNDERLINE, """i""", """n""", SPIECE_UNDERLINE, """<unk>""", """,""", SPIECE_UNDERLINE, """a""", """n""", """d""", SPIECE_UNDERLINE, """t""", """h""", """i""", """s""", SPIECE_UNDERLINE, """i""", """s""", SPIECE_UNDERLINE, """f""", """a""", """l""", """s""", """é""", """."""] )
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = [
"""Transformers (formerly known as pytorch-transformers and pytorch-pretrained-bert) provides """
"""general-purpose architectures (BERT, GPT, RoBERTa, XLM, DistilBert, XLNet...) for Natural """
"""Language Understanding (NLU) and Natural Language Generation (NLG) with over thirty-two pretrained """
"""models in one hundred plus languages and deep interoperability between Jax, PyTorch and TensorFlow.""",
"""BERT is designed to pre-train deep bidirectional representations from unlabeled text by jointly """
"""conditioning on both left and right context in all layers.""",
"""The quick brown fox jumps over the lazy dog.""",
]
# fmt: off
snake_case_ = {
"""input_ids""": [
[4, 32, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 64, 19, 8, 13, 18, 5, 13, 15, 22, 4, 28, 9, 8, 20, 9, 4, 7, 12, 4, 24, 22, 6, 8, 13, 17, 11, 39, 6, 13, 7, 9, 12, 19, 8, 13, 18, 5, 13, 12, 4, 7, 9, 14, 4, 24, 22, 6, 8, 13, 17, 11, 39, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 39, 25, 5, 13, 6, 63, 4, 24, 13, 8, 27, 10, 14, 5, 12, 4, 21, 5, 9, 5, 13, 7, 15, 39, 24, 16, 13, 24, 8, 12, 5, 4, 7, 13, 17, 11, 10, 6, 5, 17, 6, 16, 13, 5, 12, 4, 64, 40, 47, 54, 32, 23, 4, 53, 49, 32, 23, 4, 54, 8, 40, 47, 54, 32, 7, 23, 4, 69, 52, 43, 23, 4, 51, 10, 12, 6, 10, 15, 40, 5, 13, 6, 23, 4, 69, 52, 48, 5, 6, 26, 26, 26, 63, 4, 19, 8, 13, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 61, 9, 14, 5, 13, 12, 6, 7, 9, 14, 10, 9, 21, 4, 64, 48, 52, 61, 63, 4, 7, 9, 14, 4, 48, 7, 6, 16, 13, 7, 15, 4, 52, 7, 9, 21, 16, 7, 21, 5, 4, 53, 5, 9, 5, 13, 7, 6, 10, 8, 9, 4, 64, 48, 52, 53, 63, 4, 20, 10, 6, 11, 4, 8, 27, 5, 13, 4, 6, 11, 10, 13, 6, 22, 39, 6, 20, 8, 4, 24, 13, 5, 6, 13, 7, 10, 9, 5, 14, 4, 18, 8, 14, 5, 15, 12, 4, 10, 9, 4, 8, 9, 5, 4, 11, 16, 9, 14, 13, 5, 14, 4, 24, 15, 16, 12, 4, 15, 7, 9, 21, 16, 7, 21, 5, 12, 4, 7, 9, 14, 4, 14, 5, 5, 24, 4, 10, 9, 6, 5, 13, 8, 24, 5, 13, 7, 25, 10, 15, 10, 6, 22, 4, 25, 5, 6, 20, 5, 5, 9, 4, 58, 7, 37, 23, 4, 49, 22, 32, 8, 13, 17, 11, 4, 7, 9, 14, 4, 32, 5, 9, 12, 8, 13, 55, 15, 8, 20, 26, 2],
[4, 40, 47, 54, 32, 4, 10, 12, 4, 14, 5, 12, 10, 21, 9, 5, 14, 4, 6, 8, 4, 24, 13, 5, 39, 6, 13, 7, 10, 9, 4, 14, 5, 5, 24, 4, 25, 10, 14, 10, 13, 5, 17, 6, 10, 8, 9, 7, 15, 4, 13, 5, 24, 13, 5, 12, 5, 9, 6, 7, 6, 10, 8, 9, 12, 4, 19, 13, 8, 18, 4, 16, 9, 15, 7, 25, 5, 15, 5, 14, 4, 6, 5, 37, 6, 4, 25, 22, 4, 46, 8, 10, 9, 6, 15, 22, 4, 17, 8, 9, 14, 10, 6, 10, 8, 9, 10, 9, 21, 4, 8, 9, 4, 25, 8, 6, 11, 4, 15, 5, 19, 6, 4, 7, 9, 14, 4, 13, 10, 21, 11, 6, 4, 17, 8, 9, 6, 5, 37, 6, 4, 10, 9, 4, 7, 15, 15, 4, 15, 7, 22, 5, 13, 12, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[4, 32, 11, 5, 4, 45, 16, 10, 17, 28, 4, 25, 13, 8, 20, 9, 4, 19, 8, 37, 4, 46, 16, 18, 24, 12, 4, 8, 27, 5, 13, 4, 6, 11, 5, 4, 15, 7, 57, 22, 4, 14, 8, 21, 26, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
],
"""attention_mask""": [
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 0, 0, 0, 0, 0, 0, 0, 0],
[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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, 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],
]
}
# fmt: on
self.tokenizer_integration_test_util(
expected_encoding=UpperCAmelCase_ , model_name="""microsoft/speecht5_asr""" , revision="""c5ef64c71905caeccde0e4462ef3f9077224c524""" , sequences=UpperCAmelCase_ , )
| 347 | 0 |
'''simple docstring'''
# DISCLAIMER: This file is strongly influenced by https://github.com/ermongroup/ddim
from dataclasses import dataclass
from typing import Optional, Tuple, Union
import flax
import jax
import jax.numpy as jnp
from ..configuration_utils import ConfigMixin, register_to_config
from .scheduling_utils_flax import (
CommonSchedulerState,
FlaxKarrasDiffusionSchedulers,
FlaxSchedulerMixin,
FlaxSchedulerOutput,
add_noise_common,
get_velocity_common,
)
@flax.struct.dataclass
class lowerCAmelCase__ :
lowerCAmelCase_ = 42
# setable values
lowerCAmelCase_ = 42
lowerCAmelCase_ = 42
lowerCAmelCase_ = None
@classmethod
def _snake_case ( cls , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE ):
"""simple docstring"""
return cls(common=__SCREAMING_SNAKE_CASE , init_noise_sigma=__SCREAMING_SNAKE_CASE , timesteps=__SCREAMING_SNAKE_CASE )
@dataclass
class lowerCAmelCase__ ( lowerCamelCase_ ):
lowerCAmelCase_ = 42
class lowerCAmelCase__ ( lowerCamelCase_ , lowerCamelCase_ ):
lowerCAmelCase_ = [e.name for e in FlaxKarrasDiffusionSchedulers]
lowerCAmelCase_ = 42
@property
def _snake_case ( self ):
"""simple docstring"""
return True
@register_to_config
def __init__( self , __SCREAMING_SNAKE_CASE = 10_00 , __SCREAMING_SNAKE_CASE = 0.0_001 , __SCREAMING_SNAKE_CASE = 0.02 , __SCREAMING_SNAKE_CASE = "linear" , __SCREAMING_SNAKE_CASE = None , __SCREAMING_SNAKE_CASE = "fixed_small" , __SCREAMING_SNAKE_CASE = True , __SCREAMING_SNAKE_CASE = "epsilon" , __SCREAMING_SNAKE_CASE = jnp.floataa , ):
"""simple docstring"""
lowercase_ : Dict = dtype
def _snake_case ( self , __SCREAMING_SNAKE_CASE = None ):
"""simple docstring"""
if common is None:
lowercase_ : Tuple = CommonSchedulerState.create(self )
# standard deviation of the initial noise distribution
lowercase_ : Union[str, Any] = jnp.array(1.0 , dtype=self.dtype )
lowercase_ : List[Any] = jnp.arange(0 , self.config.num_train_timesteps ).round()[::-1]
return DDPMSchedulerState.create(
common=__SCREAMING_SNAKE_CASE , init_noise_sigma=__SCREAMING_SNAKE_CASE , timesteps=__SCREAMING_SNAKE_CASE , )
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE = None ):
"""simple docstring"""
return sample
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE = () ):
"""simple docstring"""
lowercase_ : Optional[Any] = self.config.num_train_timesteps // num_inference_steps
# creates integer timesteps by multiplying by ratio
# rounding to avoid issues when num_inference_step is power of 3
lowercase_ : int = (jnp.arange(0 , __SCREAMING_SNAKE_CASE ) * step_ratio).round()[::-1]
return state.replace(
num_inference_steps=__SCREAMING_SNAKE_CASE , timesteps=__SCREAMING_SNAKE_CASE , )
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE=None , __SCREAMING_SNAKE_CASE=None ):
"""simple docstring"""
lowercase_ : List[Any] = state.common.alphas_cumprod[t]
lowercase_ : str = jnp.where(t > 0 , state.common.alphas_cumprod[t - 1] , jnp.array(1.0 , dtype=self.dtype ) )
# For t > 0, compute predicted variance βt (see formula (6) and (7) from https://arxiv.org/pdf/2006.11239.pdf)
# and sample from it to get previous sample
# x_{t-1} ~ N(pred_prev_sample, variance) == add variance to pred_sample
lowercase_ : int = (1 - alpha_prod_t_prev) / (1 - alpha_prod_t) * state.common.betas[t]
if variance_type is None:
lowercase_ : str = self.config.variance_type
# hacks - were probably added for training stability
if variance_type == "fixed_small":
lowercase_ : int = jnp.clip(__SCREAMING_SNAKE_CASE , a_min=1E-2_0 )
# for rl-diffuser https://arxiv.org/abs/2205.09991
elif variance_type == "fixed_small_log":
lowercase_ : List[str] = jnp.log(jnp.clip(__SCREAMING_SNAKE_CASE , a_min=1E-2_0 ) )
elif variance_type == "fixed_large":
lowercase_ : List[Any] = state.common.betas[t]
elif variance_type == "fixed_large_log":
# Glide max_log
lowercase_ : List[Any] = jnp.log(state.common.betas[t] )
elif variance_type == "learned":
return predicted_variance
elif variance_type == "learned_range":
lowercase_ : Optional[Any] = variance
lowercase_ : Union[str, Any] = state.common.betas[t]
lowercase_ : Union[str, Any] = (predicted_variance + 1) / 2
lowercase_ : Any = frac * max_log + (1 - frac) * min_log
return variance
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE = None , __SCREAMING_SNAKE_CASE = True , ):
"""simple docstring"""
lowercase_ : Optional[int] = timestep
if key is None:
lowercase_ : int = jax.random.PRNGKey(0 )
if model_output.shape[1] == sample.shape[1] * 2 and self.config.variance_type in ["learned", "learned_range"]:
lowercase_ , lowercase_ : Optional[Any] = jnp.split(__SCREAMING_SNAKE_CASE , sample.shape[1] , axis=1 )
else:
lowercase_ : int = None
# 1. compute alphas, betas
lowercase_ : Any = state.common.alphas_cumprod[t]
lowercase_ : Optional[int] = jnp.where(t > 0 , state.common.alphas_cumprod[t - 1] , jnp.array(1.0 , dtype=self.dtype ) )
lowercase_ : int = 1 - alpha_prod_t
lowercase_ : str = 1 - alpha_prod_t_prev
# 2. compute predicted original sample from predicted noise also called
# "predicted x_0" of formula (15) from https://arxiv.org/pdf/2006.11239.pdf
if self.config.prediction_type == "epsilon":
lowercase_ : Tuple = (sample - beta_prod_t ** 0.5 * model_output) / alpha_prod_t ** 0.5
elif self.config.prediction_type == "sample":
lowercase_ : Any = model_output
elif self.config.prediction_type == "v_prediction":
lowercase_ : List[Any] = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
else:
raise ValueError(
F'''prediction_type given as {self.config.prediction_type} must be one of `epsilon`, `sample` '''
''' for the FlaxDDPMScheduler.''' )
# 3. Clip "predicted x_0"
if self.config.clip_sample:
lowercase_ : Optional[Any] = jnp.clip(__SCREAMING_SNAKE_CASE , -1 , 1 )
# 4. Compute coefficients for pred_original_sample x_0 and current sample x_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
lowercase_ : List[Any] = (alpha_prod_t_prev ** 0.5 * state.common.betas[t]) / beta_prod_t
lowercase_ : Optional[Any] = state.common.alphas[t] ** 0.5 * beta_prod_t_prev / beta_prod_t
# 5. Compute predicted previous sample µ_t
# See formula (7) from https://arxiv.org/pdf/2006.11239.pdf
lowercase_ : Optional[Any] = pred_original_sample_coeff * pred_original_sample + current_sample_coeff * sample
# 6. Add noise
def random_variance():
lowercase_ : str = jax.random.split(__SCREAMING_SNAKE_CASE , num=1 )
lowercase_ : List[Any] = jax.random.normal(__SCREAMING_SNAKE_CASE , shape=model_output.shape , dtype=self.dtype )
return (self._get_variance(__SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , predicted_variance=__SCREAMING_SNAKE_CASE ) ** 0.5) * noise
lowercase_ : Optional[Any] = jnp.where(t > 0 , random_variance() , jnp.zeros(model_output.shape , dtype=self.dtype ) )
lowercase_ : Any = pred_prev_sample + variance
if not return_dict:
return (pred_prev_sample, state)
return FlaxDDPMSchedulerOutput(prev_sample=__SCREAMING_SNAKE_CASE , state=__SCREAMING_SNAKE_CASE )
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , ):
"""simple docstring"""
return add_noise_common(state.common , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
def _snake_case ( self , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , ):
"""simple docstring"""
return get_velocity_common(state.common , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE , __SCREAMING_SNAKE_CASE )
def __len__( self ):
"""simple docstring"""
return self.config.num_train_timesteps
| 93 |
"""simple docstring"""
import datasets
__SCREAMING_SNAKE_CASE : Tuple = '\\n@InProceedings{conneau2018xnli,\n author = "Conneau, Alexis\n and Rinott, Ruty\n and Lample, Guillaume\n and Williams, Adina\n and Bowman, Samuel R.\n and Schwenk, Holger\n and Stoyanov, Veselin",\n title = "XNLI: Evaluating Cross-lingual Sentence Representations",\n booktitle = "Proceedings of the 2018 Conference on Empirical Methods\n in Natural Language Processing",\n year = "2018",\n publisher = "Association for Computational Linguistics",\n location = "Brussels, Belgium",\n}\n'
__SCREAMING_SNAKE_CASE : Dict = '\\nXNLI is a subset of a few thousand examples from MNLI which has been translated\ninto a 14 different languages (some low-ish resource). As with MNLI, the goal is\nto predict textual entailment (does sentence A imply/contradict/neither sentence\nB) and is a classification task (given two sentences, predict one of three\nlabels).\n'
__SCREAMING_SNAKE_CASE : List[str] = '\nComputes XNLI score which is just simple accuracy.\nArgs:\n predictions: Predicted labels.\n references: Ground truth labels.\nReturns:\n \'accuracy\': accuracy\nExamples:\n\n >>> predictions = [0, 1]\n >>> references = [0, 1]\n >>> xnli_metric = datasets.load_metric("xnli")\n >>> results = xnli_metric.compute(predictions=predictions, references=references)\n >>> print(results)\n {\'accuracy\': 1.0}\n'
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> List[Any]:
return (preds == labels).mean()
@datasets.utils.file_utils.add_start_docstrings(_DESCRIPTION , _KWARGS_DESCRIPTION)
class __A (datasets.Metric):
'''simple docstring'''
def lowerCAmelCase ( self : str ) ->Any:
"""simple docstring"""
return datasets.MetricInfo(
description=_DESCRIPTION , citation=_CITATION , inputs_description=_KWARGS_DESCRIPTION , features=datasets.Features(
{
"""predictions""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
"""references""": datasets.Value("""int64""" if self.config_name != """sts-b""" else """float32""" ),
} ) , codebase_urls=[] , reference_urls=[] , format="""numpy""" , )
def lowerCAmelCase ( self : Dict , UpperCAmelCase_ : Optional[int] , UpperCAmelCase_ : Any ) ->int:
"""simple docstring"""
return {"accuracy": simple_accuracy(UpperCAmelCase_ , UpperCAmelCase_ )}
| 347 | 0 |
from typing import TYPE_CHECKING
from ....utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available, is_vision_available
snake_case : int = {'''configuration_van''': ['''VAN_PRETRAINED_CONFIG_ARCHIVE_MAP''', '''VanConfig''']}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
snake_case : Dict = [
'''VAN_PRETRAINED_MODEL_ARCHIVE_LIST''',
'''VanForImageClassification''',
'''VanModel''',
'''VanPreTrainedModel''',
]
if TYPE_CHECKING:
from .configuration_van import VAN_PRETRAINED_CONFIG_ARCHIVE_MAP, VanConfig
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_van import (
VAN_PRETRAINED_MODEL_ARCHIVE_LIST,
VanForImageClassification,
VanModel,
VanPreTrainedModel,
)
else:
import sys
snake_case : Dict = _LazyModule(__name__, globals()['''__file__'''], _import_structure)
| 94 |
"""simple docstring"""
from ..utils import DummyObject, requires_backends
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Tuple ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : int ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Dict ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Union[str, Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[Any] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : Optional[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : str , **UpperCAmelCase_ : int ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : List[Any] ) ->Tuple:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Optional[int] , **UpperCAmelCase_ : Union[str, Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : List[Any] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[Any] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : List[str] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Tuple = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : Tuple , **UpperCAmelCase_ : Optional[Any] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Optional[Any] ) ->int:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : Optional[Any] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Dict ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : List[str] , **UpperCAmelCase_ : List[str] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Optional[Any] , **UpperCAmelCase_ : List[Any] ) ->Union[str, Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : Any , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[Any] ) ->List[str]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : int , *UpperCAmelCase_ : str , **UpperCAmelCase_ : Union[str, Any] ) ->List[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[Any] = ["""sentencepiece"""]
def __init__( self : List[Any] , *UpperCAmelCase_ : int , **UpperCAmelCase_ : Optional[int] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Union[str, Any] = ["""sentencepiece"""]
def __init__( self : Union[str, Any] , *UpperCAmelCase_ : Any , **UpperCAmelCase_ : str ) ->Optional[int]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Optional[int] = ["""sentencepiece"""]
def __init__( self : Tuple , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Optional[int] ) ->Dict:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Dict = ["""sentencepiece"""]
def __init__( self : Optional[int] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : List[str] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: int = ["""sentencepiece"""]
def __init__( self : Dict , *UpperCAmelCase_ : Union[str, Any] , **UpperCAmelCase_ : Optional[int] ) ->Any:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: List[str] = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : Union[str, Any] ) ->Optional[Any]:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
class __A (metaclass=snake_case__):
'''simple docstring'''
__lowercase: Any = ["""sentencepiece"""]
def __init__( self : List[str] , *UpperCAmelCase_ : List[Any] , **UpperCAmelCase_ : Optional[int] ) ->str:
"""simple docstring"""
requires_backends(self , ["""sentencepiece"""] )
| 347 | 0 |
def _A ( SCREAMING_SNAKE_CASE : Any ):
"""simple docstring"""
a__ : Any =[0] * len(SCREAMING_SNAKE_CASE )
a__ : List[Any] =[]
a__ : Dict =[1] * len(SCREAMING_SNAKE_CASE )
for values in graph.values():
for i in values:
indegree[i] += 1
for i in range(len(SCREAMING_SNAKE_CASE ) ):
if indegree[i] == 0:
queue.append(SCREAMING_SNAKE_CASE )
while queue:
a__ : int =queue.pop(0 )
for x in graph[vertex]:
indegree[x] -= 1
if long_dist[vertex] + 1 > long_dist[x]:
a__ : str =long_dist[vertex] + 1
if indegree[x] == 0:
queue.append(SCREAMING_SNAKE_CASE )
print(max(SCREAMING_SNAKE_CASE ) )
# Adjacency list of Graph
UpperCAmelCase : Union[str, Any] = {0: [2, 3, 4], 1: [2, 7], 2: [5], 3: [5, 7], 4: [7], 5: [6], 6: [7], 7: []}
longest_distance(graph)
| 95 |
"""simple docstring"""
import warnings
from ...utils import logging
from .image_processing_mobilevit import MobileViTImageProcessor
__SCREAMING_SNAKE_CASE : int = logging.get_logger(__name__)
class __A (snake_case__):
'''simple docstring'''
def __init__( self : str , *UpperCAmelCase_ : Dict , **UpperCAmelCase_ : int ) ->None:
"""simple docstring"""
warnings.warn(
"""The class MobileViTFeatureExtractor is deprecated and will be removed in version 5 of Transformers."""
""" Please use MobileViTImageProcessor instead.""" , UpperCAmelCase_ , )
super().__init__(*UpperCAmelCase_ , **UpperCAmelCase_ )
| 347 | 0 |
"""simple docstring"""
from typing import TYPE_CHECKING
from ...utils import OptionalDependencyNotAvailable, _LazyModule, is_torch_available
lowercase__ = {
"""configuration_time_series_transformer""": [
"""TIME_SERIES_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP""",
"""TimeSeriesTransformerConfig""",
],
}
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
lowercase__ = [
"""TIME_SERIES_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST""",
"""TimeSeriesTransformerForPrediction""",
"""TimeSeriesTransformerModel""",
"""TimeSeriesTransformerPreTrainedModel""",
]
if TYPE_CHECKING:
from .configuration_time_series_transformer import (
TIME_SERIES_TRANSFORMER_PRETRAINED_CONFIG_ARCHIVE_MAP,
TimeSeriesTransformerConfig,
)
try:
if not is_torch_available():
raise OptionalDependencyNotAvailable()
except OptionalDependencyNotAvailable:
pass
else:
from .modeling_time_series_transformer import (
TIME_SERIES_TRANSFORMER_PRETRAINED_MODEL_ARCHIVE_LIST,
TimeSeriesTransformerForPrediction,
TimeSeriesTransformerModel,
TimeSeriesTransformerPreTrainedModel,
)
else:
import sys
lowercase__ = _LazyModule(__name__, globals()["""__file__"""], _import_structure, module_spec=__spec__) | 96 |
"""simple docstring"""
import argparse
import json
from pathlib import Path
import requests
import timm
import torch
from huggingface_hub import hf_hub_download
from PIL import Image
from transformers import DeiTImageProcessor, ViTConfig, ViTForImageClassification, ViTImageProcessor, ViTModel
from transformers.utils import logging
logging.set_verbosity_info()
__SCREAMING_SNAKE_CASE : Tuple = logging.get_logger(__name__)
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Any:
snake_case_ = []
for i in range(config.num_hidden_layers ):
# encoder layers: output projection, 2 feedforward neural networks and 2 layernorms
rename_keys.append((f"""blocks.{i}.norm1.weight""", f"""vit.encoder.layer.{i}.layernorm_before.weight""") )
rename_keys.append((f"""blocks.{i}.norm1.bias""", f"""vit.encoder.layer.{i}.layernorm_before.bias""") )
rename_keys.append((f"""blocks.{i}.attn.proj.weight""", f"""vit.encoder.layer.{i}.attention.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.attn.proj.bias""", f"""vit.encoder.layer.{i}.attention.output.dense.bias""") )
rename_keys.append((f"""blocks.{i}.norm2.weight""", f"""vit.encoder.layer.{i}.layernorm_after.weight""") )
rename_keys.append((f"""blocks.{i}.norm2.bias""", f"""vit.encoder.layer.{i}.layernorm_after.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.weight""", f"""vit.encoder.layer.{i}.intermediate.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc1.bias""", f"""vit.encoder.layer.{i}.intermediate.dense.bias""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.weight""", f"""vit.encoder.layer.{i}.output.dense.weight""") )
rename_keys.append((f"""blocks.{i}.mlp.fc2.bias""", f"""vit.encoder.layer.{i}.output.dense.bias""") )
# projection layer + position embeddings
rename_keys.extend(
[
("""cls_token""", """vit.embeddings.cls_token"""),
("""patch_embed.proj.weight""", """vit.embeddings.patch_embeddings.projection.weight"""),
("""patch_embed.proj.bias""", """vit.embeddings.patch_embeddings.projection.bias"""),
("""pos_embed""", """vit.embeddings.position_embeddings"""),
] )
if base_model:
# layernorm + pooler
rename_keys.extend(
[
("""norm.weight""", """layernorm.weight"""),
("""norm.bias""", """layernorm.bias"""),
("""pre_logits.fc.weight""", """pooler.dense.weight"""),
("""pre_logits.fc.bias""", """pooler.dense.bias"""),
] )
# if just the base model, we should remove "vit" from all keys that start with "vit"
snake_case_ = [(pair[0], pair[1][4:]) if pair[1].startswith("""vit""" ) else pair for pair in rename_keys]
else:
# layernorm + classification head
rename_keys.extend(
[
("""norm.weight""", """vit.layernorm.weight"""),
("""norm.bias""", """vit.layernorm.bias"""),
("""head.weight""", """classifier.weight"""),
("""head.bias""", """classifier.bias"""),
] )
return rename_keys
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE=False ) -> Tuple:
for i in range(config.num_hidden_layers ):
if base_model:
snake_case_ = """"""
else:
snake_case_ = """vit."""
# read in weights + bias of input projection layer (in timm, this is a single matrix + bias)
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.weight""" )
snake_case_ = state_dict.pop(f"""blocks.{i}.attn.qkv.bias""" )
# next, add query, keys and values (in that order) to the state dict
snake_case_ = in_proj_weight[
: config.hidden_size, :
]
snake_case_ = in_proj_bias[: config.hidden_size]
snake_case_ = in_proj_weight[
config.hidden_size : config.hidden_size * 2, :
]
snake_case_ = in_proj_bias[
config.hidden_size : config.hidden_size * 2
]
snake_case_ = in_proj_weight[
-config.hidden_size :, :
]
snake_case_ = in_proj_bias[-config.hidden_size :]
def _a ( _SCREAMING_SNAKE_CASE ) -> List[str]:
snake_case_ = ["""head.weight""", """head.bias"""]
for k in ignore_keys:
state_dict.pop(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> str:
snake_case_ = dct.pop(_SCREAMING_SNAKE_CASE )
snake_case_ = val
def _a ( ) -> Any:
snake_case_ = """http://images.cocodataset.org/val2017/000000039769.jpg"""
snake_case_ = Image.open(requests.get(_SCREAMING_SNAKE_CASE , stream=_SCREAMING_SNAKE_CASE ).raw )
return im
@torch.no_grad()
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> Any:
snake_case_ = ViTConfig()
snake_case_ = False
# dataset (ImageNet-21k only or also fine-tuned on ImageNet 2012), patch_size and image_size
if vit_name[-5:] == "in21k":
snake_case_ = True
snake_case_ = int(vit_name[-12:-10] )
snake_case_ = int(vit_name[-9:-6] )
else:
snake_case_ = 1_000
snake_case_ = """huggingface/label-files"""
snake_case_ = """imagenet-1k-id2label.json"""
snake_case_ = json.load(open(hf_hub_download(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , repo_type="""dataset""" ) , """r""" ) )
snake_case_ = {int(_SCREAMING_SNAKE_CASE ): v for k, v in idalabel.items()}
snake_case_ = idalabel
snake_case_ = {v: k for k, v in idalabel.items()}
snake_case_ = int(vit_name[-6:-4] )
snake_case_ = int(vit_name[-3:] )
# size of the architecture
if "deit" in vit_name:
if vit_name[9:].startswith("""tiny""" ):
snake_case_ = 192
snake_case_ = 768
snake_case_ = 12
snake_case_ = 3
elif vit_name[9:].startswith("""small""" ):
snake_case_ = 384
snake_case_ = 1_536
snake_case_ = 12
snake_case_ = 6
else:
pass
else:
if vit_name[4:].startswith("""small""" ):
snake_case_ = 768
snake_case_ = 2_304
snake_case_ = 8
snake_case_ = 8
elif vit_name[4:].startswith("""base""" ):
pass
elif vit_name[4:].startswith("""large""" ):
snake_case_ = 1_024
snake_case_ = 4_096
snake_case_ = 24
snake_case_ = 16
elif vit_name[4:].startswith("""huge""" ):
snake_case_ = 1_280
snake_case_ = 5_120
snake_case_ = 32
snake_case_ = 16
# load original model from timm
snake_case_ = timm.create_model(_SCREAMING_SNAKE_CASE , pretrained=_SCREAMING_SNAKE_CASE )
timm_model.eval()
# load state_dict of original model, remove and rename some keys
snake_case_ = timm_model.state_dict()
if base_model:
remove_classification_head_(_SCREAMING_SNAKE_CASE )
snake_case_ = create_rename_keys(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
for src, dest in rename_keys:
rename_key(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
read_in_q_k_v(_SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE )
# load HuggingFace model
if vit_name[-5:] == "in21k":
snake_case_ = ViTModel(_SCREAMING_SNAKE_CASE ).eval()
else:
snake_case_ = ViTForImageClassification(_SCREAMING_SNAKE_CASE ).eval()
model.load_state_dict(_SCREAMING_SNAKE_CASE )
# Check outputs on an image, prepared by ViTImageProcessor/DeiTImageProcessor
if "deit" in vit_name:
snake_case_ = DeiTImageProcessor(size=config.image_size )
else:
snake_case_ = ViTImageProcessor(size=config.image_size )
snake_case_ = image_processor(images=prepare_img() , return_tensors="""pt""" )
snake_case_ = encoding["""pixel_values"""]
snake_case_ = model(_SCREAMING_SNAKE_CASE )
if base_model:
snake_case_ = timm_model.forward_features(_SCREAMING_SNAKE_CASE )
assert timm_pooled_output.shape == outputs.pooler_output.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.pooler_output , atol=1E-3 )
else:
snake_case_ = timm_model(_SCREAMING_SNAKE_CASE )
assert timm_logits.shape == outputs.logits.shape
assert torch.allclose(_SCREAMING_SNAKE_CASE , outputs.logits , atol=1E-3 )
Path(_SCREAMING_SNAKE_CASE ).mkdir(exist_ok=_SCREAMING_SNAKE_CASE )
print(f"""Saving model {vit_name} to {pytorch_dump_folder_path}""" )
model.save_pretrained(_SCREAMING_SNAKE_CASE )
print(f"""Saving image processor to {pytorch_dump_folder_path}""" )
image_processor.save_pretrained(_SCREAMING_SNAKE_CASE )
if __name__ == "__main__":
__SCREAMING_SNAKE_CASE : int = argparse.ArgumentParser()
# Required parameters
parser.add_argument(
'--vit_name',
default='vit_base_patch16_224',
type=str,
help='Name of the ViT timm model you\'d like to convert.',
)
parser.add_argument(
'--pytorch_dump_folder_path', default=None, type=str, help='Path to the output PyTorch model directory.'
)
__SCREAMING_SNAKE_CASE : int = parser.parse_args()
convert_vit_checkpoint(args.vit_name, args.pytorch_dump_folder_path)
| 347 | 0 |
'''simple docstring'''
def a ( __a , __a , __a=False ) -> str:
'''simple docstring'''
if isinstance(__a , __a ) and isinstance(__a , __a ):
UpperCamelCase__ :List[str] = len(set_a.intersection(__a ) )
if alternative_union:
UpperCamelCase__ :Optional[Any] = len(__a ) + len(__a )
else:
UpperCamelCase__ :Union[str, Any] = len(set_a.union(__a ) )
return intersection / union
if isinstance(__a , (list, tuple) ) and isinstance(__a , (list, tuple) ):
UpperCamelCase__ :Tuple = [element for element in set_a if element in set_b]
if alternative_union:
UpperCamelCase__ :Dict = len(__a ) + len(__a )
return len(__a ) / union
else:
UpperCamelCase__ :List[Any] = set_a + [element for element in set_b if element not in set_a]
return len(__a ) / len(__a )
return len(__a ) / len(__a )
return None
if __name__ == "__main__":
__snake_case = {'''a''', '''b''', '''c''', '''d''', '''e'''}
__snake_case = {'''c''', '''d''', '''e''', '''f''', '''h''', '''i'''}
print(jaccard_similarity(set_a, set_b)) | 97 |
"""simple docstring"""
import unittest
import numpy as np
from transformers import RoFormerConfig, is_flax_available
from transformers.testing_utils import require_flax, slow
from ...test_modeling_flax_common import FlaxModelTesterMixin, ids_tensor, random_attention_mask
if is_flax_available():
import jax.numpy as jnp
from transformers.models.roformer.modeling_flax_roformer import (
FlaxRoFormerForMaskedLM,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerModel,
)
class __A (unittest.TestCase):
'''simple docstring'''
def __init__( self : List[Any] , UpperCAmelCase_ : Union[str, Any] , UpperCAmelCase_ : Tuple=13 , UpperCAmelCase_ : List[Any]=7 , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : int=True , UpperCAmelCase_ : Dict=99 , UpperCAmelCase_ : str=32 , UpperCAmelCase_ : Tuple=5 , UpperCAmelCase_ : Union[str, Any]=4 , UpperCAmelCase_ : Any=37 , UpperCAmelCase_ : int="gelu" , UpperCAmelCase_ : Any=0.1 , UpperCAmelCase_ : List[str]=0.1 , UpperCAmelCase_ : Dict=512 , UpperCAmelCase_ : Optional[Any]=16 , UpperCAmelCase_ : Dict=2 , UpperCAmelCase_ : str=0.02 , UpperCAmelCase_ : str=4 , ) ->Tuple:
"""simple docstring"""
snake_case_ = parent
snake_case_ = batch_size
snake_case_ = seq_length
snake_case_ = is_training
snake_case_ = use_attention_mask
snake_case_ = use_token_type_ids
snake_case_ = use_labels
snake_case_ = vocab_size
snake_case_ = hidden_size
snake_case_ = num_hidden_layers
snake_case_ = num_attention_heads
snake_case_ = intermediate_size
snake_case_ = hidden_act
snake_case_ = hidden_dropout_prob
snake_case_ = attention_probs_dropout_prob
snake_case_ = max_position_embeddings
snake_case_ = type_vocab_size
snake_case_ = type_sequence_label_size
snake_case_ = initializer_range
snake_case_ = num_choices
def lowerCAmelCase ( self : Optional[int] ) ->str:
"""simple docstring"""
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.vocab_size )
snake_case_ = None
if self.use_attention_mask:
snake_case_ = random_attention_mask([self.batch_size, self.seq_length] )
snake_case_ = None
if self.use_token_type_ids:
snake_case_ = ids_tensor([self.batch_size, self.seq_length] , self.type_vocab_size )
snake_case_ = RoFormerConfig(
vocab_size=self.vocab_size , hidden_size=self.hidden_size , num_hidden_layers=self.num_hidden_layers , num_attention_heads=self.num_attention_heads , intermediate_size=self.intermediate_size , hidden_act=self.hidden_act , hidden_dropout_prob=self.hidden_dropout_prob , attention_probs_dropout_prob=self.attention_probs_dropout_prob , max_position_embeddings=self.max_position_embeddings , type_vocab_size=self.type_vocab_size , is_decoder=UpperCAmelCase_ , initializer_range=self.initializer_range , )
return config, input_ids, token_type_ids, attention_mask
def lowerCAmelCase ( self : List[str] ) ->Dict:
"""simple docstring"""
snake_case_ = self.prepare_config_and_inputs()
snake_case_ , snake_case_ , snake_case_ , snake_case_ = config_and_inputs
snake_case_ = {"""input_ids""": input_ids, """token_type_ids""": token_type_ids, """attention_mask""": attention_mask}
return config, inputs_dict
@require_flax
class __A (snake_case__ , unittest.TestCase):
'''simple docstring'''
__lowercase: Union[str, Any] = True
__lowercase: int = (
(
FlaxRoFormerModel,
FlaxRoFormerForMaskedLM,
FlaxRoFormerForSequenceClassification,
FlaxRoFormerForTokenClassification,
FlaxRoFormerForMultipleChoice,
FlaxRoFormerForQuestionAnswering,
)
if is_flax_available()
else ()
)
def lowerCAmelCase ( self : Optional[Any] ) ->Tuple:
"""simple docstring"""
snake_case_ = FlaxRoFormerModelTester(self )
@slow
def lowerCAmelCase ( self : Any ) ->List[str]:
"""simple docstring"""
for model_class_name in self.all_model_classes:
snake_case_ = model_class_name.from_pretrained("""junnyu/roformer_chinese_small""" , from_pt=UpperCAmelCase_ )
snake_case_ = model(np.ones((1, 1) ) )
self.assertIsNotNone(UpperCAmelCase_ )
@require_flax
class __A (unittest.TestCase):
'''simple docstring'''
@slow
def lowerCAmelCase ( self : str ) ->Dict:
"""simple docstring"""
snake_case_ = FlaxRoFormerForMaskedLM.from_pretrained("""junnyu/roformer_chinese_base""" )
snake_case_ = jnp.array([[0, 1, 2, 3, 4, 5]] )
snake_case_ = model(UpperCAmelCase_ )[0]
snake_case_ = 50_000
snake_case_ = (1, 6, vocab_size)
self.assertEqual(output.shape , UpperCAmelCase_ )
snake_case_ = jnp.array(
[[[-0.1_205, -1.0_265, 0.2_922], [-1.5_134, 0.1_974, 0.1_519], [-5.0_135, -3.9_003, -0.8_404]]] )
self.assertTrue(jnp.allclose(output[:, :3, :3] , UpperCAmelCase_ , atol=1E-4 ) )
| 347 | 0 |
"""simple docstring"""
import requests
from bsa import BeautifulSoup
def a_ ( lowerCamelCase , lowerCamelCase ):
UpperCAmelCase__ = BeautifulSoup(requests.get(lowerCamelCase , params=lowerCamelCase ).content , 'html.parser' )
UpperCAmelCase__ = soup.find('div' , attrs={'class': 'gs_ri'} )
UpperCAmelCase__ = div.find('div' , attrs={'class': 'gs_fl'} ).find_all('a' )
return anchors[2].get_text()
if __name__ == "__main__":
lowerCAmelCase__ : Optional[int] = {
'title': (
'Precisely geometry controlled microsupercapacitors for ultrahigh areal '
'capacitance, volumetric capacitance, and energy density'
),
'journal': 'Chem. Mater.',
'volume': 30,
'pages': '3979-3990',
'year': 2_018,
'hl': 'en',
}
print(get_citation('https://scholar.google.com/scholar_lookup', params=params))
| 98 |
"""simple docstring"""
from __future__ import annotations
def _a ( _SCREAMING_SNAKE_CASE , _SCREAMING_SNAKE_CASE ) -> bool:
snake_case_ = get_failure_array(_SCREAMING_SNAKE_CASE )
# 2) Step through text searching for pattern
snake_case_ , snake_case_ = 0, 0 # index into text, pattern
while i < len(_SCREAMING_SNAKE_CASE ):
if pattern[j] == text[i]:
if j == (len(_SCREAMING_SNAKE_CASE ) - 1):
return True
j += 1
# if this is a prefix in our pattern
# just go back far enough to continue
elif j > 0:
snake_case_ = failure[j - 1]
continue
i += 1
return False
def _a ( _SCREAMING_SNAKE_CASE ) -> list[int]:
snake_case_ = [0]
snake_case_ = 0
snake_case_ = 1
while j < len(_SCREAMING_SNAKE_CASE ):
if pattern[i] == pattern[j]:
i += 1
elif i > 0:
snake_case_ = failure[i - 1]
continue
j += 1
failure.append(_SCREAMING_SNAKE_CASE )
return failure
if __name__ == "__main__":
# Test 1)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abc1abc12'
__SCREAMING_SNAKE_CASE : Optional[int] = 'alskfjaldsabc1abc1abc12k23adsfabcabc'
__SCREAMING_SNAKE_CASE : List[str] = 'alskfjaldsk23adsfabcabc'
assert kmp(pattern, texta) and not kmp(pattern, texta)
# Test 2)
__SCREAMING_SNAKE_CASE : int = 'ABABX'
__SCREAMING_SNAKE_CASE : Optional[Any] = 'ABABZABABYABABX'
assert kmp(pattern, text)
# Test 3)
__SCREAMING_SNAKE_CASE : Any = 'AAAB'
__SCREAMING_SNAKE_CASE : List[Any] = 'ABAAAAAB'
assert kmp(pattern, text)
# Test 4)
__SCREAMING_SNAKE_CASE : Optional[int] = 'abcdabcy'
__SCREAMING_SNAKE_CASE : str = 'abcxabcdabxabcdabcdabcy'
assert kmp(pattern, text)
# Test 5)
__SCREAMING_SNAKE_CASE : Any = 'aabaabaaa'
assert get_failure_array(pattern) == [0, 1, 0, 1, 2, 3, 4, 5, 2]
| 347 | 0 |
import numpy as np
from numpy import ndarray
from scipy.optimize import Bounds, LinearConstraint, minimize
def A_ ( A__ ) -> float:
return np.dot(A__ , A__ )
class A__ :
"""simple docstring"""
def __init__( self , *,
lowercase = np.inf , lowercase = "linear" , lowercase = 0.0 , ) -> None:
'''simple docstring'''
a__ : Tuple = regularization
a__ : Optional[Any] = gamma
if kernel == "linear":
a__ : Optional[Any] = self.__linear
elif kernel == "rbf":
if self.gamma == 0:
raise ValueError('rbf kernel requires gamma')
if not isinstance(self.gamma , (float, int)):
raise ValueError('gamma must be float or int')
if not self.gamma > 0:
raise ValueError('gamma must be > 0')
a__ : str = self.__rbf
# in the future, there could be a default value like in sklearn
# sklear: def_gamma = 1/(n_features * X.var()) (wiki)
# previously it was 1/(n_features)
else:
a__ : Optional[int] = F'Unknown kernel: {kernel}'
raise ValueError(lowercase)
def __lowercase ( self , lowercase , lowercase) -> float:
'''simple docstring'''
return np.dot(lowercase , lowercase)
def __lowercase ( self , lowercase , lowercase) -> float:
'''simple docstring'''
return np.exp(-(self.gamma * norm_squared(vectora - vectora)))
def __lowercase ( self , lowercase , lowercase) -> None:
'''simple docstring'''
a__ : List[str] = observations
a__ : Dict = classes
# using Wolfe's Dual to calculate w.
# Primal problem: minimize 1/2*norm_squared(w)
# constraint: yn(w . xn + b) >= 1
#
# With l a vector
# Dual problem: maximize sum_n(ln) -
# 1/2 * sum_n(sum_m(ln*lm*yn*ym*xn . xm))
# constraint: self.C >= ln >= 0
# and sum_n(ln*yn) = 0
# Then we get w using w = sum_n(ln*yn*xn)
# At the end we can get b ~= mean(yn - w . xn)
#
# Since we use kernels, we only need l_star to calculate b
# and to classify observations
((a__) , ) : Optional[int] = np.shape(lowercase)
def to_minimize(lowercase) -> float:
a__ : Tuple = 0
((a__) , ) : Optional[int] = np.shape(lowercase)
for i in range(lowercase):
for j in range(lowercase):
s += (
candidate[i]
* candidate[j]
* classes[i]
* classes[j]
* self.kernel(observations[i] , observations[j])
)
return 1 / 2 * s - sum(lowercase)
a__ : Optional[Any] = LinearConstraint(lowercase , 0 , 0)
a__ : str = Bounds(0 , self.regularization)
a__ : List[str] = minimize(
lowercase , np.ones(lowercase) , bounds=lowercase , constraints=[ly_contraint]).x
a__ : Dict = l_star
# calculating mean offset of separation plane to points
a__ : int = 0
for i in range(lowercase):
for j in range(lowercase):
s += classes[i] - classes[i] * self.optimum[i] * self.kernel(
observations[i] , observations[j])
a__ : List[str] = s / n
def __lowercase ( self , lowercase) -> int:
'''simple docstring'''
a__ : int = sum(
self.optimum[n]
* self.classes[n]
* self.kernel(self.observations[n] , lowercase)
for n in range(len(self.classes)))
return 1 if s + self.offset >= 0 else -1
if __name__ == "__main__":
import doctest
doctest.testmod()
| 99 |
"""simple docstring"""
from transformers import BertTokenizer, EncoderDecoderModel, SeqaSeqTrainer, SeqaSeqTrainingArguments
from transformers.testing_utils import TestCasePlus, require_torch, slow
from transformers.utils import is_datasets_available
if is_datasets_available():
import datasets
class __A (snake_case__):
'''simple docstring'''
@slow
@require_torch
def lowerCAmelCase ( self : Union[str, Any] ) ->Dict:
"""simple docstring"""
snake_case_ = EncoderDecoderModel.from_encoder_decoder_pretrained("""prajjwal1/bert-tiny""" , """prajjwal1/bert-tiny""" )
snake_case_ = BertTokenizer.from_pretrained("""bert-base-uncased""" )
snake_case_ = bertabert.config.encoder.vocab_size
snake_case_ = tokenizer.sep_token_id
snake_case_ = tokenizer.cls_token_id
snake_case_ = 128
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""train[:1%]""" )
snake_case_ = datasets.load_dataset("""cnn_dailymail""" , """3.0.0""" , split="""validation[:1%]""" )
snake_case_ = train_dataset.select(range(32 ) )
snake_case_ = val_dataset.select(range(16 ) )
snake_case_ = 4
def _map_to_encoder_decoder_inputs(UpperCAmelCase_ : int ):
# Tokenizer will automatically set [BOS] <text> [EOS]
snake_case_ = tokenizer(batch["""article"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=512 )
snake_case_ = tokenizer(batch["""highlights"""] , padding="""max_length""" , truncation=UpperCAmelCase_ , max_length=128 )
snake_case_ = inputs.input_ids
snake_case_ = inputs.attention_mask
snake_case_ = outputs.input_ids
snake_case_ = outputs.input_ids.copy()
snake_case_ = [
[-100 if token == tokenizer.pad_token_id else token for token in labels] for labels in batch["""labels"""]
]
snake_case_ = outputs.attention_mask
assert all(len(UpperCAmelCase_ ) == 512 for x in inputs.input_ids )
assert all(len(UpperCAmelCase_ ) == 128 for x in outputs.input_ids )
return batch
def _compute_metrics(UpperCAmelCase_ : Union[str, Any] ):
snake_case_ = pred.label_ids
snake_case_ = pred.predictions
# all unnecessary tokens are removed
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = tokenizer.batch_decode(UpperCAmelCase_ , skip_special_tokens=UpperCAmelCase_ )
snake_case_ = sum([int(pred_str[i] == label_str[i] ) for i in range(len(UpperCAmelCase_ ) )] ) / len(UpperCAmelCase_ )
return {"accuracy": accuracy}
# map train dataset
snake_case_ = train_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
train_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
# same for validation dataset
snake_case_ = val_dataset.map(
_map_to_encoder_decoder_inputs , batched=UpperCAmelCase_ , batch_size=UpperCAmelCase_ , remove_columns=["""article""", """highlights"""] , )
val_dataset.set_format(
type="""torch""" , columns=["""input_ids""", """attention_mask""", """decoder_input_ids""", """decoder_attention_mask""", """labels"""] , )
snake_case_ = self.get_auto_remove_tmp_dir()
snake_case_ = SeqaSeqTrainingArguments(
output_dir=UpperCAmelCase_ , per_device_train_batch_size=UpperCAmelCase_ , per_device_eval_batch_size=UpperCAmelCase_ , predict_with_generate=UpperCAmelCase_ , evaluation_strategy="""steps""" , do_train=UpperCAmelCase_ , do_eval=UpperCAmelCase_ , warmup_steps=0 , eval_steps=2 , logging_steps=2 , )
# instantiate trainer
snake_case_ = SeqaSeqTrainer(
model=UpperCAmelCase_ , args=UpperCAmelCase_ , compute_metrics=_compute_metrics , train_dataset=UpperCAmelCase_ , eval_dataset=UpperCAmelCase_ , tokenizer=UpperCAmelCase_ , )
# start training
trainer.train()
| 347 | 0 |
"""simple docstring"""
from typing import Dict
import numpy as np
from ..utils import add_end_docstrings, is_tf_available, is_torch_available, logging
from .base import PIPELINE_INIT_ARGS, GenericTensor, Pipeline, PipelineException
if is_tf_available():
import tensorflow as tf
from ..tf_utils import stable_softmax
if is_torch_available():
import torch
__magic_name__ = logging.get_logger(__name__)
@add_end_docstrings(
__a , R'''
top_k (`int`, defaults to 5):
The number of predictions to return.
targets (`str` or `List[str]`, *optional*):
When passed, the model will limit the scores to the passed targets instead of looking up in the whole
vocab. If the provided targets are not in the model vocab, they will be tokenized and the first resulting
token will be used (with a warning, and that might be slower).
''' , )
class SCREAMING_SNAKE_CASE_ ( __a ):
"""simple docstring"""
def snake_case_ ( self , lowerCAmelCase__):
if self.framework == "tf":
__SCREAMING_SNAKE_CASE = tf.where(input_ids == self.tokenizer.mask_token_id).numpy()
elif self.framework == "pt":
__SCREAMING_SNAKE_CASE = torch.nonzero(input_ids == self.tokenizer.mask_token_id , as_tuple=lowerCAmelCase__)
else:
raise ValueError("""Unsupported framework""")
return masked_index
def snake_case_ ( self , lowerCAmelCase__):
__SCREAMING_SNAKE_CASE = self.get_masked_index(lowerCAmelCase__)
__SCREAMING_SNAKE_CASE = np.prod(masked_index.shape)
if numel < 1:
raise PipelineException(
"""fill-mask""" , self.model.base_model_prefix , f"No mask_token ({self.tokenizer.mask_token}) found on the input" , )
def snake_case_ ( self , lowerCAmelCase__):
if isinstance(lowerCAmelCase__ , lowerCAmelCase__):
for model_input in model_inputs:
self._ensure_exactly_one_mask_token(model_input["""input_ids"""][0])
else:
for input_ids in model_inputs["input_ids"]:
self._ensure_exactly_one_mask_token(lowerCAmelCase__)
def snake_case_ ( self , lowerCAmelCase__ , lowerCAmelCase__=None , **lowerCAmelCase__):
if return_tensors is None:
__SCREAMING_SNAKE_CASE = self.framework
__SCREAMING_SNAKE_CASE = self.tokenizer(lowerCAmelCase__ , return_tensors=lowerCAmelCase__)
self.ensure_exactly_one_mask_token(lowerCAmelCase__)
return model_inputs
def snake_case_ ( self , lowerCAmelCase__):
__SCREAMING_SNAKE_CASE = self.model(**lowerCAmelCase__)
__SCREAMING_SNAKE_CASE = model_inputs["""input_ids"""]
return model_outputs
def snake_case_ ( self , lowerCAmelCase__ , lowerCAmelCase__=5 , lowerCAmelCase__=None):
# Cap top_k if there are targets
if target_ids is not None and target_ids.shape[0] < top_k:
__SCREAMING_SNAKE_CASE = target_ids.shape[0]
__SCREAMING_SNAKE_CASE = model_outputs["""input_ids"""][0]
__SCREAMING_SNAKE_CASE = model_outputs["""logits"""]
if self.framework == "tf":
__SCREAMING_SNAKE_CASE = tf.where(input_ids == self.tokenizer.mask_token_id).numpy()[:, 0]
__SCREAMING_SNAKE_CASE = outputs.numpy()
__SCREAMING_SNAKE_CASE = outputs[0, masked_index, :]
__SCREAMING_SNAKE_CASE = stable_softmax(lowerCAmelCase__ , axis=-1)
if target_ids is not None:
__SCREAMING_SNAKE_CASE = tf.gather_nd(tf.squeeze(lowerCAmelCase__ , 0) , target_ids.reshape(-1 , 1))
__SCREAMING_SNAKE_CASE = tf.expand_dims(lowerCAmelCase__ , 0)
__SCREAMING_SNAKE_CASE = tf.math.top_k(lowerCAmelCase__ , k=lowerCAmelCase__)
__SCREAMING_SNAKE_CASE ,__SCREAMING_SNAKE_CASE = topk.values.numpy(), topk.indices.numpy()
else:
__SCREAMING_SNAKE_CASE = torch.nonzero(input_ids == self.tokenizer.mask_token_id , as_tuple=lowerCAmelCase__).squeeze(-1)
# Fill mask pipeline supports only one ${mask_token} per sample
__SCREAMING_SNAKE_CASE = outputs[0, masked_index, :]
__SCREAMING_SNAKE_CASE = logits.softmax(dim=-1)
if target_ids is not None:
__SCREAMING_SNAKE_CASE = probs[..., target_ids]
__SCREAMING_SNAKE_CASE ,__SCREAMING_SNAKE_CASE = probs.topk(lowerCAmelCase__)
__SCREAMING_SNAKE_CASE = []
__SCREAMING_SNAKE_CASE = values.shape[0] == 1
for i, (_values, _predictions) in enumerate(zip(values.tolist() , predictions.tolist())):
__SCREAMING_SNAKE_CASE = []
for v, p in zip(_values , _predictions):
# Copy is important since we're going to modify this array in place
__SCREAMING_SNAKE_CASE = input_ids.numpy().copy()
if target_ids is not None:
__SCREAMING_SNAKE_CASE = target_ids[p].tolist()
__SCREAMING_SNAKE_CASE = p
# Filter padding out:
__SCREAMING_SNAKE_CASE = tokens[np.where(tokens != self.tokenizer.pad_token_id)]
# Originally we skip special tokens to give readable output.
# For multi masks though, the other [MASK] would be removed otherwise
# making the output look odd, so we add them back
__SCREAMING_SNAKE_CASE = self.tokenizer.decode(lowerCAmelCase__ , skip_special_tokens=lowerCAmelCase__)
__SCREAMING_SNAKE_CASE = {"""score""": v, """token""": p, """token_str""": self.tokenizer.decode([p]), """sequence""": sequence}
row.append(lowerCAmelCase__)
result.append(lowerCAmelCase__)
if single_mask:
return result[0]
return result
def snake_case_ ( self , lowerCAmelCase__ , lowerCAmelCase__=None):
if isinstance(lowerCAmelCase__ , lowerCAmelCase__):
__SCREAMING_SNAKE_CASE = [targets]
try:
__SCREAMING_SNAKE_CASE = self.tokenizer.get_vocab()
except Exception:
__SCREAMING_SNAKE_CASE = {}
__SCREAMING_SNAKE_CASE = []
for target in targets:
__SCREAMING_SNAKE_CASE = vocab.get(lowerCAmelCase__ , lowerCAmelCase__)
if id_ is None:
__SCREAMING_SNAKE_CASE = self.tokenizer(
lowerCAmelCase__ , add_special_tokens=lowerCAmelCase__ , return_attention_mask=lowerCAmelCase__ , return_token_type_ids=lowerCAmelCase__ , max_length=1 , truncation=lowerCAmelCase__ , )["""input_ids"""]
if len(lowerCAmelCase__) == 0:
logger.warning(
f"The specified target token `{target}` does not exist in the model vocabulary. "
"""We cannot replace it with anything meaningful, ignoring it""")
continue
__SCREAMING_SNAKE_CASE = input_ids[0]
# XXX: If users encounter this pass
# it becomes pretty slow, so let's make sure
# The warning enables them to fix the input to
# get faster performance.
logger.warning(
f"The specified target token `{target}` does not exist in the model vocabulary. "
f"Replacing with `{self.tokenizer.convert_ids_to_tokens(id_)}`.")
target_ids.append(id_)
__SCREAMING_SNAKE_CASE = list(set(lowerCAmelCase__))
if len(lowerCAmelCase__) == 0:
raise ValueError("""At least one target must be provided when passed.""")
__SCREAMING_SNAKE_CASE = np.array(lowerCAmelCase__)
return target_ids
def snake_case_ ( self , lowerCAmelCase__=None , lowerCAmelCase__=None):
__SCREAMING_SNAKE_CASE = {}
if targets is not None:
__SCREAMING_SNAKE_CASE = self.get_target_ids(lowerCAmelCase__ , lowerCAmelCase__)
__SCREAMING_SNAKE_CASE = target_ids
if top_k is not None:
__SCREAMING_SNAKE_CASE = top_k
if self.tokenizer.mask_token_id is None:
raise PipelineException(
"""fill-mask""" , self.model.base_model_prefix , """The tokenizer does not define a `mask_token`.""")
return {}, {}, postprocess_params
def __call__( self , lowerCAmelCase__ , *lowerCAmelCase__ , **lowerCAmelCase__):
__SCREAMING_SNAKE_CASE = super().__call__(lowerCAmelCase__ , **lowerCAmelCase__)
if isinstance(lowerCAmelCase__ , lowerCAmelCase__) and len(lowerCAmelCase__) == 1:
return outputs[0]
return outputs
| 100 |
"""simple docstring"""
# this script reports modified .py files under the desired list of top-level sub-dirs passed as a list of arguments, e.g.:
# python ./utils/get_modified_files.py utils src tests examples
#
# it uses git to find the forking point and which files were modified - i.e. files not under git won't be considered
# since the output of this script is fed into Makefile commands it doesn't print a newline after the results
import re
import subprocess
import sys
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output('git merge-base main HEAD'.split()).decode('utf-8')
__SCREAMING_SNAKE_CASE : Tuple = subprocess.check_output(f"""git diff --name-only {fork_point_sha}""".split()).decode('utf-8').split()
__SCREAMING_SNAKE_CASE : Any = '|'.join(sys.argv[1:])
__SCREAMING_SNAKE_CASE : Optional[Any] = re.compile(Rf"""^({joined_dirs}).*?\.py$""")
__SCREAMING_SNAKE_CASE : List[str] = [x for x in modified_files if regex.match(x)]
print(' '.join(relevant_modified_files), end='')
| 347 | 0 |