Persian-conversational-dataset.py

# coding=utf-8
# persian-conversational-dataset
"""TODO(empathetic_dialogues): Add a description here."""


import csv,json

import datasets
from datasets.tasks import QuestionAnsweringExtractive



logger = datasets.logging.get_logger(__name__)

_DESCRIPTION = """\
persian-conversational-dataset
"""

_URL = "https://huggingface.co/datasets/Kamtera/Persian-conversational-dataset/blob/main/"
_URLS = [
    "dadrah_dataset.json",
    "dadrah_dataset1-1000_10000.json",
    "dadrah_dataset1-10000_100000.json",
    "dadrah_dataset1-100000_276342.json",
]

class persianConversation(datasets.GeneratorBasedBuilder):

    # VERSION = datasets.Version("0.1.0")

    def _info(self):
        # TODO(empathetic_dialogues): Specifies the datasets.DatasetInfo object
        return datasets.DatasetInfo(
            # This is the description that will appear on the datasets page.
            description=_DESCRIPTION,
            # datasets.features.FeatureConnectors
            features=datasets.Features(
                {
                    "title": datasets.Value("string"),
                    "question": datasets.Value("string"),
                    "answers": datasets.Sequence(datasets.Value("string")),
                    "keywords": datasets.Sequence(datasets.Value("string")),
                    # These are the features of your dataset like images, labels ...
                }
            ),
            # If there's a common (input, target) tuple from the features,
            # specify them here. They'll be used if as_supervised=True in
            # builder.as_dataset.
            supervised_keys=None,
            
            
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        # TODO(empathetic_dialogues): Downloads the data and defines the splits
        # dl_manager is a datasets.download.DownloadManager that can be used to
        # download and extract URLs
        downloaded_files = dl_manager.download(_URLS)
        logger.info("| > downloaded files")
        logger.info(downloaded_files)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={
                    "files": downloaded_files[1:],
                    "split_file": "train",
                },
            ),
            datasets.SplitGenerator(
                name=datasets.Split.TEST,
                # These kwargs will be passed to _generate_examples
                gen_kwargs={
                    "files": downloaded_files[0:1],
                    "split_file": "test"
                },
            ),
        ]

    def _generate_examples(self, files, split_file):
        """Yields examples."""
        logger.info("| > generate examples for "+split_file)
        logger.info(files)
        for path in files:
            with open(path, 'r', encoding='utf-8') as fmm:
                data=json.load(fmm)
                for id_, row in enumerate(data):
                    title=row[0]
                    question=row[1]
                    answers=row[2]
                    keywords=row[3]
                    if id_==20:
                        break;
                    yield id_, {
                        "title": title,
                        "question": question,
                        "answers": answers,
                        "keywords": keywords,
                    }