melisa.py

import datasets
import csv

_LICENSE = """
TO DO: Licencia
"""

with open("README.md", "r") as f:
    lines = iter(f.readlines())
    for line in lines:
        if "### Dataset Summary" in line:
            break
    next(lines)
    _DESCRIPTION = next(lines)

_CITATION = """
TO DO: Cita
"""

_LANGUAGES = {
    "es": "Spanish",
    "pt": "Portuguese"
}

_ALL_LANGUAGES = "all_languages"

_VERSION = "1.0.0"

_HOMEPAGE_URL = "https://github.com/lpsc-fiuba/MeLiSA"

_DOWNLOAD_URL = "./{lang}/{split}.csv"


class MeLiSAConfig(datasets.BuilderConfig):
    """BuilderConfig for MeLiSA."""

    def __init__(self, languages=None, **kwargs):
        """Constructs a MeLiSAConfig.
        Args:
            **kwargs: keyword arguments forwarded to super.
        """
        super(MeLiSAConfig, self).__init__(version=datasets.Version(_VERSION, ""), **kwargs)
        self.languages = languages



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

    BUILDER_CONFIGS = [
        MeLiSAConfig(
            name=_ALL_LANGUAGES,
            languages=_LANGUAGES,
            description="A collection of Mercado Libre reviews specifically designed to aid research in spanish and portuguese sentiment classification.",
        )
    ] + [
        MeLiSAConfig(
            name=lang,
            languages=[lang],
            description=f"{_LANGUAGES[lang]} examples from a collection of Mercado Libre reviews specifically designed to aid research in sentiment classification",
        )
        for lang in _LANGUAGES
    ]
    BUILDER_CONFIG_CLASS = MeLiSAConfig
    DEFAULT_CONFIG_NAME = _ALL_LANGUAGES

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "country": datasets.Value("string"),
                    "category": datasets.Value("string"),
                    "review_content": datasets.Value("string"),
                    "review_title": datasets.Value("string"),
                    "review_rate": datasets.Value("int32")
                }
            ),
            supervised_keys=None,
            license=_LICENSE,
            homepage=_HOMEPAGE_URL,
            citation=_CITATION,
        )
    
    def _split_generators(self, dl_manager):
        train_urls = [_DOWNLOAD_URL.format(split="train", lang=lang) for lang in self.config.languages]
        dev_urls = [_DOWNLOAD_URL.format(split="validation", lang=lang) for lang in self.config.languages]
        test_urls = [_DOWNLOAD_URL.format(split="test", lang=lang) for lang in self.config.languages]

        train_paths = dl_manager.download_and_extract(train_urls)
        dev_paths = dl_manager.download_and_extract(dev_urls)
        test_paths = dl_manager.download_and_extract(test_urls)

        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"file_paths": train_paths}),
            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"file_paths": dev_paths}),
            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"file_paths": test_paths}),
        ]

    def _generate_examples(self, file_paths):
        """Generate features given the directory path.
        Args:
            file_path: path where the tsv file is stored
        Yields:
            The features.
        """

        for file_path in file_paths:
            with open(file_path, "r", encoding="utf-8") as csvfile:
                reader = csv.DictReader(csvfile)
                for i, row in enumerate(reader):
                    yield i, row