dutch-snli.py

import csv
import os

import datasets



_DESCRIPTION = """\
This is the Dutch version of the original SNLI dataset. The translation was performed using Google Translate. Original SNLI available at https://nlp.stanford.edu/projects/snli/
"""


class DutchSnli(datasets.GeneratorBasedBuilder):
    """The Dutch-translated Stanford Natural Language Inference (SNLI) Corpus."""

    BUILDER_CONFIGS = [
        datasets.BuilderConfig(
            name="plain_text",
            version=datasets.Version("1.0.0", ""),
            description="Plain text import of SNLI",
        )
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "premise": datasets.Value("string"),
                    "hypothesis": datasets.Value("string"),
                    "label": datasets.features.ClassLabel(names=["entailment", "neutral", "contradiction"]),
                }
            ),
            # No default supervised_keys (as we have to pass both premise
            # and hypothesis as input).
            supervised_keys=None,
            homepage="",
            citation="",
        )

    def _split_generators(self, dl_manager: datasets.DownloadManager):
        data_dir = dl_manager._base_path
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TEST, gen_kwargs={"filepath": os.path.join(data_dir, "test.tsv")}
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN, gen_kwargs={"filepath": os.path.join(data_dir, "train.tsv")}
            ),
        ]

    def _generate_examples(self, filepath):
        """This function returns the examples in the raw (text) form."""
        with open(filepath, encoding="utf-8") as f:
            reader = csv.DictReader(f, delimiter="\t", quoting=csv.QUOTE_NONE)
            for idx, row in enumerate(reader):
                if row["label"] != '-':
                    yield idx, {
                        "premise": row["premise"],
                        "hypothesis": row["hypothesis"],
                        "label": row["label"],
                    }