build_parquet.py

"""TODO(xtreme): Add a description here."""

import csv
import glob
import os

import datasets


_UD_POS_LANG = {
    "Afrikaans": 'af',
    "Arabic": 'ar',
    "Basque": 'eu',
    "Bulgarian": 'bg',
    "Dutch": 'nl',
    "English": 'en',
    "Estonian": 'et',
    "Finnish": 'fi',
    "French": 'fr',
    "German": 'de',
    "Greek": 'el',
    "Hebrew": 'he',
    "Hindi": 'hi',
    "Hungarian": 'hu',
    "Indonesian": 'id',
    "Italian": 'it',
    "Japanese": 'ja',
    "Kazakh": 'kk',
    "Korean": 'ko',
    "Chinese": 'zh',
    "Marathi": 'mr',
    "Persian": 'fa',
    "Portuguese": 'pt',
    "Russian": 'ru',
    "Spanish": 'es',
    "Tagalog": 'tl',
    "Tamil": 'ta',
    "Telugu": 'te',
    "Thai": 'th',
    "Turkish": 'tr',
    "Urdu": 'ur',
    "Vietnamese": 'vi',
    "Yoruba": 'yo',
}


_DATA_URLS = {
    "2_5": "https://lindat.mff.cuni.cz/repository/xmlui/bitstream/handle/11234/1-3105/ud-treebanks-v2.5.tgz",
    "2_7": 'https://lindat.mff.cuni.cz/repository/xmlui/bitstream/handle/11234/1-3424/ud-treebanks-v2.7.tgz',
}


def generate_examples(folder: str, lang: str, split: str):
    """Yields examples."""
    for path in glob.glob(f'{folder}/UD_{lang}*/*{split}.conllu'):
        # For lang other than [see below], we exclude Arabic-NYUAD which does not contains any words, only _
        if lang in ["Kazakh", "Tagalog", "Thai", "Yoruba"] or "NYUAD" not in path:
            print('read', path)
            with open(path) as file:
                data = csv.reader(file, delimiter="\t", quoting=csv.QUOTE_NONE)
                tokens = []
                pos_tags = []
                for id_row, row in enumerate(data):
                    if len(row) >= 10 and row[1] != "_" and row[3] != "_":
                        tokens.append(row[1])
                        pos_tags.append(row[3])
                    if len(row) == 0 and len(tokens) > 0:
                        yield {"tokens": tokens, "pos_tags": pos_tags}
                        tokens = []
                        pos_tags = []


def main():
    features = datasets.Features({
        "tokens": datasets.Sequence(datasets.Value("string")),
        "pos_tags": datasets.Sequence(datasets.features.ClassLabel(names=[
            "ADJ",
            "ADP",
            "ADV",
            "AUX",
            "CCONJ",
            "DET",
            "INTJ",
            "NOUN",
            "NUM",
            "PART",
            "PRON",
            "PROPN",
            "PUNCT",
            "SCONJ",
            "SYM",
            "VERB",
            "X",
        ])),
    })

    path = 'ud-treebanks-v2.7/'
    if '2.7' in path:  # xtreme-r
        _UD_POS_LANG.update({
            'Lithuanian': 'lt', 'Polish': 'pl', 'Ukrainian': 'uk', 'Wolof': 'wo', 'Romanian': 'ro',
        })
    for lang, code in _UD_POS_LANG.items():
        os.makedirs(f'{path}/parquet/{code}/', exist_ok=True)
        splits = ['test'] if code != 'en' else ['train', 'dev', 'test']
        for split in splits:
            ds = datasets.Dataset.from_generator(
                generate_examples, features=features, keep_in_memory=True, gen_kwargs=dict(
                    folder=path, lang=lang, split=split
                )
            )
            sp = f'{path}/parquet/{code}/{split}.parquet'
            ds.to_parquet(sp)
            print('save', sp)


if __name__ == '__main__':
    main()