_unsilence_voc.py

# coding=utf-8
# Copyright 2022 HuggingFace Datasets Authors.
#
# 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.

# Lint as: python3
"""UnSilenceVOC dataset"""


import datasets
import re
from datasets import ClassLabel, Sequence, Value


_CITATION = """\
TODO
"""

_DESCRIPTION = """\
TODO
"""

NE_MAIN_LABELS = [
    "B-Organization",
    "B-Organization,B-Place",
    "B-Organization,I-Person",
    "B-Organization,I-Place",
    "B-Person",
    "B-Person,B-Place",
    "B-Person,I-Place",
    "B-Place",
    "I-Organization",
    "I-Organization,B-Place",
    "I-Organization,I-Person",
    "I-Organization,I-Person,B-Place",
    "I-Organization,I-Person,I-Place",
    "I-Organization,I-Place",
    "I-Person",
    "I-Person,B-Place",
    "I-Person,I-Place",
    "I-Place",
    "O",
]

NE_PER_NAME = ["I-ProperName", "O", "B-ProperName", ""]
NE_PER_GENDER = [
    "B-Group",
    "B-Man",
    "B-Man,B-Unspecified",
    "B-Man,I-Woman",
    "B-Unspecified",
    "B-Unspecified,I-Woman",
    "B-Woman",
    "I-Group",
    "I-Man",
    "I-Man,I-Unspecified",
    "I-Man,I-Woman",
    "I-Unspecified",
    "I-Unspecified,I-Woman",
    "I-Woman",
    "NE-PER-GENDER",
    "O",
]

NE_PER_LEGAL_STATUS = [
    "B-Enslaved",
    "B-Freed",
    "B-Unspecified",
    "I-Enslaved",
    "I-Freed",
    "I-Unspecified",
    "NE-PER-LEGAL-STATUS",
    "O",
]

NE_PER_ROLE = [
    "B-Acting_Notary",
    "B-Beneficiary",
    "B-Notary",
    "B-Other",
    "B-Testator",
    "B-Testator_Beneficiary",
    "B-Witness",
    "I-Acting_Notary",
    "I-Beneficiary",
    "I-Beneficiary,B-Other",
    "I-Beneficiary,I-Other",
    "I-Notary",
    "I-Other",
    "I-Testator",
    "I-Testator_Beneficiary",
    "I-Witness",
    "NE-PER-ROLE",
    "O",
]

NE_ORG_BENEFICIARY = [
    "B-No",
    "B-Yes",
    "I-No",
    "I-Yes",
    "NE-ORG-BENEFICIARY",
    "O",
]

_BASE_URL = (
    "https://raw.githubusercontent.com/budh333/UnSilence_VOC/main/processed_data"
)

_URLS = {
    "train": f"{_BASE_URL}/train-nl.tsv",
    "test": f"{_BASE_URL}/test-nl.tsv",
    "dev": f"{_BASE_URL}/dev-nl.tsv",
}


class UnSilenceVOC(datasets.GeneratorBasedBuilder):
    """UnSilence VOC dataset."""

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "tokens": Sequence(datasets.Value("string")),
                    "NE-MAIN": Sequence(
                        ClassLabel(names=NE_MAIN_LABELS),
                    ),
                    "NE-PER-NAME": Sequence(ClassLabel(names=NE_PER_NAME)),
                    "NE-PER-GENDER": Sequence(ClassLabel(names=NE_PER_GENDER)),
                    "NE-PER-LEGAL-STATUS": Sequence(
                        ClassLabel(names=NE_PER_LEGAL_STATUS)
                    ),
                    "NE-PER-ROLE": Sequence(ClassLabel(names=NE_PER_ROLE)),
                    "NE-ORG-BENEFICIARY": Sequence(
                        ClassLabel(names=NE_ORG_BENEFICIARY)
                    ),
                    "MISC": Value("string"),
                    "document_id": datasets.Value("string"),
                }
            ),
            homepage="TODO",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        downloaded_files = dl_manager.download_and_extract(_URLS)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filepath": downloaded_files["train"]},
            )
        ]

    def _generate_examples(self, filepath):
        document_id_re = re.compile(r"# document_path = ..(\/.*.txt)")
        with open(filepath, encoding="utf-8") as f:
            guid = 0
            tokens = []
            NE_MAIN_LABELS = []
            NE_PER_NAME = []
            NE_PER_GENDER = []
            NE_PER_LEGAL_STATUS = []
            NE_PER_ROLE = []
            NE_ORG_BENEFICIARY = []
            MISC = []

            for line in f:
                if line.startswith("TOKEN"):
                    continue
                if line.startswith("#") or line.startswith("\t"):
                    document_id_match = re.search(document_id_re, line)
                    if document_id_match:
                        document_id = document_id_match.groups(0)[0]

                    if not tokens:
                        continue
                    yield guid, {
                        "tokens": tokens,
                        "NE-MAIN": NE_MAIN_LABELS,
                        "NE-PER-NAME": NE_PER_NAME,
                        "NE-PER-GENDER": NE_PER_GENDER,
                        "NE-PER-LEGAL-STATUS": NE_PER_LEGAL_STATUS,
                        "NE-PER-ROLE": NE_PER_ROLE,
                        "NE-ORG-BENEFICIARY": NE_ORG_BENEFICIARY,
                        "MISC": MISC,
                        "document_id": document_id,
                    }
                    guid += 1
                    tokens = []
                    NE_MAIN_LABELS = []
                    NE_PER_NAME = []
                    NE_PER_GENDER = []
                    NE_PER_LEGAL_STATUS = []
                    NE_PER_ROLE = []
                    NE_ORG_BENEFICIARY = []
                    MISC = []
                else:
                    # tokens are tab separated
                    splits = line.split("\t")
                    tokens.append(splits[0])
                    NE_MAIN_LABELS.append(splits[1])
                    NE_PER_NAME.append(splits[2])
                    NE_PER_GENDER.append(splits[3])
                    NE_PER_LEGAL_STATUS.append(splits[4])
                    NE_PER_ROLE.append(splits[5])
                    NE_ORG_BENEFICIARY.append(splits[6])
                    MISC.append(splits[-1].replace("\n", ""))

            # last example
            yield guid, {
                "tokens": tokens,
                "NE-MAIN": NE_MAIN_LABELS,
                "NE-PER-NAME": NE_PER_NAME,
                "NE-PER-GENDER": NE_PER_GENDER,
                "NE-PER-LEGAL-STATUS": NE_PER_LEGAL_STATUS,
                "NE-PER-ROLE": NE_PER_ROLE,
                "NE-ORG-BENEFICIARY": NE_ORG_BENEFICIARY,
                "MISC": MISC,
                "document_id": "document_id",
            }