personachat_truecased.py

import json

import datasets
from datasets.features import Sequence


_URLS = {
    "full": {
        "train": "./personachat_truecased_full_train.json",
        "valid": "./personachat_truecased_full_valid.json"
    },
    "sample": {
        "train": "./personachat_truecased_sample_train.json",
        "valid": "./personachat_truecased_sample_valid.json"
    }
}

_DESCRIPTION = """\
A version of the PersonaChat dataset that has been true-cased, and also has been given more normalized punctuation.
The original PersonaChat dataset is in all lower case, and has extra space around each clause/sentence separating
punctuation mark. This version of the dataset has more of a natural language look, with sentence capitalization,
proper noun capitalization, and normalized whitespace. Also, each dialogue turn includes a pool of distractor
candidate responses, which can be used by a multiple choice regularization loss during training.
"""

_CITATION = """\
@article{zhang2018personalizing,
  title={Personalizing dialogue agents: I have a dog, do you have pets too?},
  author={Zhang, Saizheng and Dinan, Emily and Urbanek, Jack and Szlam, Arthur and Kiela, Douwe and Weston, Jason},
  journal={arXiv preprint arXiv:1801.07243},
  year={2018}
}
"""

_LICENSE = "Like the original PersonaChat dataset, this dataset is released under the CC BY 4.0 license."


class PersonachatTruecased(datasets.GeneratorBasedBuilder):
    """
    Version of the PersonaChat dataset that includes true-casing, normalized punctuation, and candidate distractor
    responses for each dialogue turn, for including a multiple choice regularzation loss while training.
    """

    VERSION = datasets.Version("1.0.0")

    BUILDER_CONFIGS = [
        datasets.BuilderConfig(name="full", version=VERSION, description="The full dataset."),
        datasets.BuilderConfig(
            name="sample", version=VERSION, description="A sample sample of the dataset, useful for testing."
        )
    ]

    DEFAULT_CONFIG_NAME = "full"

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features({
                "personality": Sequence(datasets.Value("string")),
                "candidates": Sequence(datasets.Value("string")),
                "history": Sequence(datasets.Value("string")),
                "conv_id": datasets.Value("int32"),
                "utterance_idx": datasets.Value("int32")
            }),
            citation=_CITATION,
            license=_LICENSE
        )

    def _split_generators(self, dl_manager: datasets.DownloadManager):
        split_paths = dl_manager.download(_URLS[self.config.name])
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={"data_path": split_paths["train"]}
            ),
            datasets.SplitGenerator(
                name=datasets.Split.VALIDATION,
                gen_kwargs={"data_path": split_paths["valid"]}
            )
        ]

    def _generate_examples(self, data_path: str):
        with open(data_path) as f:
            data = json.load(f)
        for conv_id, conv in enumerate(data):
            personality = conv["personality"]
            for utterance_idx, utterance in enumerate(conv["utterances"]):
                id_ = f"{conv_id}-{utterance_idx}"
                yield id_, {
                    "personality": personality,
                    "candidates": utterance["candidates"],
                    "history": utterance["history"],
                    "conv_id": conv_id,
                    "utterance_idx": utterance_idx
                }