quakeset.py

# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# 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 json
import os

import datasets
import h5py
import numpy as np

# Find for instance the citation on arxiv or on the dataset repo/website
_CITATION = """\
WIP
"""

# You can copy an official description
_DESCRIPTION = """\
QuakeSet is a dataset of earthquake images from the Copernicus Sentinel-1 satellites. 
It contains images from before, after an earthquake, and a sample before the "before" sample.
Ground truth contains magnitudes and locations of earthquakes provided by ISC.
"""

_HOMEPAGE = "https://huggingface.co/datasets/DarthReca/quakeset"

_LICENSE = "OPENRAIL"

# The HuggingFace Datasets library doesn't host the datasets but only points to the original files.
# This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method)
_URLS = "earthquakes.h5"


class QuakeSet(datasets.GeneratorBasedBuilder):
    """TODO: Short description of my dataset."""

    VERSION = datasets.Version("1.0.0")

    # This is an example of a dataset with multiple configurations.
    # If you don't want/need to define several sub-sets in your dataset,
    # just remove the BUILDER_CONFIG_CLASS and the BUILDER_CONFIGS attributes.

    # If you need to make complex sub-parts in the datasets with configurable options
    # You can create your own builder configuration class to store attribute, inheriting from datasets.BuilderConfig
    # BUILDER_CONFIG_CLASS = MyBuilderConfig

    # You will be able to load one or the other configurations in the following list with
    # data = datasets.load_dataset('my_dataset', 'first_domain')
    # data = datasets.load_dataset('my_dataset', 'second_domain')
    BUILDER_CONFIGS = [
        datasets.BuilderConfig(
            name="default",
            version=VERSION,
            description="Default configuration. No other configuration is available",
        )
    ]

    DEFAULT_CONFIG_NAME = "default"  # It's not mandatory to have a default configuration. Just use one if it make sense.

    def _info(self):
        features = datasets.Features(
            {
                "pre_post_image": datasets.Array3D(
                    shape=(512, 513, 2), dtype="float32"
                ),
                "affected": datasets.ClassLabel(num_classes=2),
                "magnitude": datasets.Value("float32"),
                "hypocenter": datasets.Sequence(datasets.Value("float32"), length=3),
                "epsg": datasets.Value("int32"),
                "x": datasets.Sequence(datasets.Value("float32"), length=512),
                "y": datasets.Sequence(datasets.Value("float32"), length=512),
            }
        )

        return datasets.DatasetInfo(
            # This is the description that will appear on the datasets page.
            description=_DESCRIPTION,
            # This defines the different columns of the dataset and their types
            features=features,  # Here we define them above because they are different between the two configurations
            # If there's a common (input, target) tuple from the features, uncomment supervised_keys line below and
            # specify them. They'll be used if as_supervised=True in builder.as_dataset.
            # supervised_keys=("sentence", "label"),
            # Homepage of the dataset for documentation
            homepage=_HOMEPAGE,
            # License for the dataset if available
            license=_LICENSE,
            # Citation for the dataset
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        # If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name
        # dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLS
        # It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files.
        # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive
        urls = _URLS
        data_dir = dl_manager.download(urls)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={
                    "filepath": os.path.join(data_dir, "earthquakes.h5"),
                    "split": "train",
                },
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={
                    "filepath": os.path.join(data_dir, "earthquakes.h5"),
                    "split": "validation",
                },
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={
                    "filepath": os.path.join(data_dir, "earthquakes.h5"),
                    "split": "test",
                },
            ),
        ]

    # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
    def _generate_examples(self, filepath, split):
        # The `key` is for legacy reasons (tfds) and is not important in itself, but must be unique for each example.
        sample_ids = []
        with h5py.File(filepath) as f:
            for key, patches in f.items():
                attributes = dict(f[key].attrs)
                if attributes["split"] != split:
                    continue
                sample_ids += [(f"{key}/{p}", 1, attributes) for p in patches.keys()]
                sample_ids += [
                    (f"{key}/{p}", 0, attributes)
                    for p, v in patches.items()
                    if "before" in v
                ]

            for sample_id, label, attributes in sample_ids:
                if "x" in sample_id or "y" in sample_id:
                    continue

                resource_id, patch_id = sample_id.split("/")
                x = f[resource_id]["x"][...]
                y = f[resource_id]["y"][...]
                x_start = int(patch_id.split("_")[1]) % (x.shape[0] // 512)
                y_start = int(patch_id.split("_")[1]) // (x.shape[0] // 512)
                x = x[x_start * 512 : (x_start + 1) * 512]
                y = y[y_start * 512 : (y_start + 1) * 512]

                pre_key = "pre" if label == 1 else "before"
                post_key = "post" if label == 1 else "pre"
                pre_sample = f[sample_id][pre_key][...]
                post_sample = f[sample_id][post_key][...]
                pre_sample = np.nan_to_num(pre_sample, nan=0).transpose(2, 0, 1)
                post_sample = np.nan_to_num(post_sample, nan=0).transpose(2, 0, 1)
                sample = np.concatenate(
                    [pre_sample, post_sample], axis=0, dtype=np.float32
                )

                yield f"{sample_id}/{post_key}", {
                    "pre_post_image": sample,
                    "affected": label,
                    "magnitude": np.float32(attributes["magnitude"]),
                    "hypocenter": attributes["hypocenter"],
                    "epsg": attributes["epsg"],
                    "x": x.flatten(),
                    "y": y.flatten(),
                }