defamation-japanese-twitter.py

# coding=utf-8
# Copyright 2020 The TensorFlow Datasets Authors and the 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


import json

import datasets


logger = datasets.logging.get_logger(__name__)


_CITATION = """
"""

_DESCRIPTION = """
"""
_URL = "https://raw.githubusercontent.com/kubotaissei/defamation_japanese_twitter/master/input/"
_URLS = {"dataset": _URL + "data_twitter.json"}


class DefamationJapaneseTwitterConfig(datasets.BuilderConfig):
    """BuilderConfig for DefamationJapaneseTwitterConfig."""

    def __init__(self, **kwargs):
        """BuilderConfig for DefamationJapaneseTwitterConfig.

        Args:
          **kwargs: keyword arguments forwarded to super.
        """
        super(DefamationJapaneseTwitterConfig, self).__init__(**kwargs)


class DefamationJapaneseTwitter(datasets.GeneratorBasedBuilder):
    """DefamationJapaneseTwitterConfig: Japanese Defamation Detection Twitter Dataset. Version 1.0."""

    BUILDER_CONFIGS = [
        DefamationJapaneseTwitterConfig(
            name="plain_text",
            version=datasets.Version("1.0.0", ""),
            description="Plain text",
        ),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "target": datasets.features.Sequence(
                        datasets.ClassLabel(names=["C", "A1", "A2", "A3"])
                    ),
                    "label": datasets.features.Sequence(
                        datasets.ClassLabel(names=["C", "B1", "B2", "B3", "B4"])
                    ),
                    "user_id_list": datasets.features.Sequence(datasets.Value("int32")),
                }
            ),
            # No default supervised_keys (as we have to pass both question
            # and context as input).
            supervised_keys=None,
            homepage="",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        downloaded_files = dl_manager.download_and_extract(_URLS)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filepath": downloaded_files},
            ),
        ]

    def _generate_examples(self, filepath):
        """This function returns the examples depending on split"""

        with open(filepath["dataset"], encoding="utf-8") as f:
            data = json.load(f)

        for i, (id, v) in enumerate(data.items()):
            yield i, {
                "id": id,
                "target": v["target"],
                "label": v["label"],
                "user_id_list": v["user_id_list"],
            }