new_dataset.py

import csv
import json
import os
import datasets
import pandas as pd

_CITATION = """
@article{alam2021review,
  title={A Review of Bangla Natural Language Processing Tasks and the Utility of Transformer Models},
  author={Alam, Firoj and Hasan, Md Arid and Alam, Tanvir and Khan, Akib and Tajrin, Janntatul and Khan, Naira and Chowdhury, Shammur Absar},
  journal={arXiv preprint arXiv:2107.03844},
  year={2021}
}
@article{alam2020bangla,
  title={Bangla Text Classification using Transformers},
  author={Alam, Tanvirul and Khan, Akib and Alam, Firoj},
  journal={arXiv preprint arXiv:2011.04446},
  year={2020}
}

@article{kunchukuttan2020ai4bharat,
 author = {Anoop Kunchukuttan and Divyanshu Kakwani and Satish Golla and Gokul N.C. and Avik Bhattacharyya and Mitesh M. Khapra and Pratyush Kumar},
 journal = {arXiv preprint arXiv:2005.00085},
 title = {AI4Bharat-IndicNLP Corpus: Monolingual Corpora and Word Embeddings for Indic Languages},
 year = {2020}
}
"""

_DESCRIPTION = """
The dataset contains six different class labels for news categorization task and is available with training, development, and test splits with 11,284, 1,411, and 1,411 news articles, respectively.
"""
_HOMEPAGE = "https://github.com/banglanlp/bnlp-resources"

_LICENCE = "https://creativecommons.org/licenses/by-nc-sa/4.0/"

_URLS = {
    "https://github.com/banglanlp/bnlp-resources/tree/main/news_categorization"
}


class new_datasetConfig(datasets.BuilderConfig):
    def __init__(self, features, data_url, citation, url, label_classes=("False", "True"), **kwargs):
        super(new_datasetConfig, self).__init__(version=datasets.Version("0.1.1"), **kwargs)
        self.features = features
        self.label_classes = label_classes
        self.data_url = data_url
        self.citation = citation
        self.url = url


class new_dataset(datasets.GeneratorBasedBuilder):
    VERSION = datasets.Version("0.1.1")

    BUILDER_CONFIGS = [
        new_datasetConfig(
            name="new_dataset",
            features=["text", "class_label"],
            data_url="https://github.com/banglanlp/bnlp-resources/tree/main/news_categorization",
            citation=_CITATION,
            url="https://github.com/google"
        )
    ]
    print(BUILDER_CONFIGS[0].name)

    def _info(self):
        features = datasets.Features(
            {
                "text": datasets.Values("string"),
                "class_label": datasets.Values("string"),

            }
        )
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            supervised_keys=None,
            homepage=_HOMEPAGE,
            license=_LICENCE,
            citation=_CITATION,

        )


def _Split_genertors(self, dl_manager):
    urls = _URLS[self.config.name]
    print(urls)
    data_dir = "./data"
    return [
        datasets._Split_genertor(
            name=datasets.Split.TRAIN,
            gen_kwargs={
                "filepath": os.path.join(data_dir, "train.tsv"),
                "split": "train",
            },
        ),
        datasets.SplitGenerator(
            name=datasets.Split.TEST,
            gen_kwargs={
                "filepath": os.path.join(data_dir, "test.tsv"),
                "split": "test",
            },
        ),
        datasets.SplitGenerator(
            name=datasets.Split.VALIDATION,
            gen_kwargs={
                "filepath": os.path.join(data_dir, "dev.tsv"),
                "split": "dev",
            },
        ),
    ]


def _generate_examples(self, filepath, split):
    with open(filepath, encoding="utf-8") as f:
        reader = csv.reader(f, delimiter="\t")
        next(reader)
        for row in reader:
            # data = csv.loads(row)
            yield row[0], {
                "text": row[1],
                "class_label": row[2],
            }