political-advertising-pl.py

from pathlib import Path

import datasets
import pandas as pd

logger = datasets.logging.get_logger(__name__)

_CITATION = """\
@inproceedings{augustyniak-etal-2020-political,
    title = "Political Advertising Dataset: the use case of the Polish 2020 Presidential Elections",
    author = "Augustyniak, Lukasz  and
      Rajda, Krzysztof  and
      Kajdanowicz, Tomasz  and
      Bernaczyk, Micha{\l}",
    booktitle = "Proceedings of the The Fourth Widening Natural Language Processing Workshop",
    month = jul,
    year = "2020",
    address = "Seattle, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.winlp-1.28",
    pages = "110--114"
}
"""

_DESCRIPTION = "Polish Political Advertising Dataset"

_HOMEPAGE = "https://github.com/laugustyniak/misinformation"

_URLS = {
    "train": "https://huggingface.co/datasets/laugustyniak/political-advertising-pl/resolve/main/train.parquet",
    "test": "https://huggingface.co/datasets/laugustyniak/political-advertising-pl/resolve/main/test.parquet",
    "validation": "https://huggingface.co/datasets/laugustyniak/political-advertising-pl/resolve/main/dev.parquet",
}

DATA_PATH = Path(".")


class PoliticalAdvertisingConfig(datasets.BuilderConfig):
    def __init__(self, **kwargs):
        super(PoliticalAdvertisingConfig, self).__init__(**kwargs)


class PoliticalAdvertisingDataset(datasets.GeneratorBasedBuilder):
    VERSION = datasets.Version("1.0.0")

    BUILDER_CONFIGS = [
        datasets.BuilderConfig(name="political-advertising-pl", version=VERSION)
    ]

    def _info(self):
        features = datasets.Features(
            {
                "id": datasets.Value("string"),
                "tokens": datasets.Sequence(datasets.Value("string")),
                "tags": datasets.Sequence(
                    datasets.features.ClassLabel(
                        names=[
                            "O",
                            "B-DEFENSE_AND_SECURITY",
                            "I-DEFENSE_AND_SECURITY",
                            "B-EDUCATION",
                            "I-EDUCATION",
                            "B-FOREIGN_POLICY",
                            "I-FOREIGN_POLICY",
                            "B-HEALHCARE",
                            "I-HEALHCARE",
                            "B-IMMIGRATION",
                            "I-IMMIGRATION",
                            "B-INFRASTRUCTURE_AND_ENVIROMENT",
                            "I-INFRASTRUCTURE_AND_ENVIROMENT",
                            "B-POLITICAL_AND_LEGAL_SYSTEM",
                            "I-POLITICAL_AND_LEGAL_SYSTEM",
                            "B-SOCIETY",
                            "I-SOCIETY",
                            "B-WELFARE",
                            "I-WELFARE",
                        ]
                    )
                ),
            }
        )

        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            supervised_keys=None,
            homepage=_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["train"]},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                gen_kwargs={"filepath": downloaded_files["test"]},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                gen_kwargs={"filepath": downloaded_files["validation"]},
            ),
        ]

    def _generate_examples(self, filepath: str):
        logger.info("⏳ Generating examples from = %s", filepath)
        df = pd.read_parquet(filepath)

        for row_id, row in df.iterrows():
            yield row_id, {
                "id": str(row_id),
                "tokens": row.tokens,
                "tags": row.tags,
            }