dataset_script.py

import json
import datasets

_DESCRIPTION = "Multi-domain dataset with separate configurations per domain."
_CITATION = "Your citation here"

class MultiDomainDataset(datasets.GeneratorBasedBuilder):
    BUILDER_CONFIGS = [
        datasets.BuilderConfig(name="domain_0", version=datasets.Version("1.0.0"), description="Domain 0 data"),
        datasets.BuilderConfig(name="domain_1", version=datasets.Version("1.0.0"), description="Domain 1 data"),
        datasets.BuilderConfig(name="domain_2", version=datasets.Version("1.0.0"), description="Domain 2 data"),
        datasets.BuilderConfig(name="domain_3", version=datasets.Version("1.0.0"), description="Domain 3 data"),
        datasets.BuilderConfig(name="domain_4", version=datasets.Version("1.0.0"), description="Domain 4 data"),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features({
                "text": datasets.Value("string"),
            }),
            supervised_keys=None,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        domain_file = f"{self.config.name}.jsonl"
        data_path = dl_manager.download_and_extract(domain_file)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"filepath": data_path},
            )
        ]

    def _generate_examples(self, filepath):
        with open(filepath, encoding="utf-8") as f:
            for idx, line in enumerate(f):
                yield idx, json.loads(line)