clef24_dataset_en.py

"""Multilang Dataset loading script."""

from datasets import DatasetDict, DatasetInfo, BuilderConfig, Version, GeneratorBasedBuilder
from datasets import SplitGenerator, Split, Features, Value

import os

_DESCRIPTION = """
This dataset includes multilingual data for language classification tasks across several languages.
"""

_CITATION = """\
@InProceedings{huggingface:multilang_dataset,
title = {Multilingual Text Dataset},
authors = {Your Name},
year = {2024}
}
"""

_LICENSE = "Your dataset's license here."

class MultilangDataset(GeneratorBasedBuilder):
    """A multilingual text dataset."""

    BUILDER_CONFIGS = [
        BuilderConfig(name="multilang_dataset", version=Version("1.0.0"), description="Multilingual dataset for text classification."),
    ]

    DEFAULT_CONFIG_NAME = "multilang_dataset"  # Default configuration name.

    def _info(self):
        return DatasetInfo(
            description=_DESCRIPTION,
            features=Features({
                "Sentence_id": Value("string"),
                "Text": Value("string"),
                "class_label": Value("string"),
            }),
            supervised_keys=("Text", "class_label"),
            homepage="https://www.example.com",
            citation=_CITATION,
            license=_LICENSE,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        # Assumes your dataset is located in "."
        data_dir = os.path.abspath(".")
        splits = {"train": Split.TRAIN, "dev": Split.VALIDATION, "dev-test": Split.TEST}
        
        return [
            SplitGenerator(
                name=splits[split],
                gen_kwargs={
                    "filepath": os.path.join(data_dir, f"{split}.tsv"),
                    "split": splits[split]
                },
            )
            for split in splits.keys()
        ]

    def _generate_examples(self, filepath, split):
        """Yields examples."""
        with open(filepath, encoding="utf-8") as f:
            for id_, row in enumerate(f):
                if id_ == 0:  # Optionally skip header
                    continue
                cols = row.strip().split('\t')
                yield f"{split}_{id_}", {
                    "sentence_id": cols[0],
                    "sentence": cols[1],
                    "label": cols[2],
                }