finstsb_dataset.py

import csv
import datasets

class Finstsb(datasets.GeneratorBasedBuilder):
    """Finstsb dataset."""

    VERSION = datasets.Version("1.0.0")

    def _info(self):
        return datasets.DatasetInfo(
            description="Finstsb dataset",
            features=datasets.Features(
                {
                    "sentence1": datasets.Value("string"),
                    "sentence2": datasets.Value("string"),
                    "gpt_score": datasets.Value("int32"),
                    "score": datasets.Value("int32"),
                }
            ),
            supervised_keys=None,
            homepage="https://huggingface.co/datasets",
            citation="",
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        urls_to_download = {
            "dev": "path/to/your/finstsb_to_dev.csv",
            "test": "path/to/your/finstsb_to_test.csv",
        }
        downloaded_files = dl_manager.download_and_extract(urls_to_download)

        return [
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                gen_kwargs={"filepath": downloaded_files["dev"]},
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                gen_kwargs={"filepath": downloaded_files["test"]},
            ),
        ]

    def _generate_examples(self, filepath):
        """Yields examples."""
        with open(filepath, encoding="utf-8") as f:
            reader = csv.DictReader(f)
            for id_, row in enumerate(reader):
                yield id_, {
                    "sentence1": row["sentence1"],
                    "sentence2": row["sentence2"],
                    "gpt_score": int(row["gpt_score"]),
                    "score": int(row["score"]),
                }