itu_faroese_danish.py

# coding=utf-8
# Copyright 2020 HuggingFace Datasets Authors.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

# Lint as: python3
"""Faroese-Danish parallel corpus from ITU Copenhagen"""

import os

import datasets


logger = datasets.logging.get_logger(__name__)


_CITATION = """\

"""

_DESCRIPTION = """\

"""

_URL = "itu_fo_da_monolithic.tsv"


class ItuFaroeseDanishConfig(datasets.BuilderConfig):
    """BuilderConfig for ItuFaroeseDanish"""

    def __init__(self, **kwargs):
        """BuilderConfig ItuFaroeseDanish.

        Args:
          **kwargs: keyword arguments forwarded to super.
        """
        super(ItuFaroeseDanishConfig, self).__init__(**kwargs)


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

    BUILDER_CONFIGS = [
        ItuFaroeseDanishConfig(name="ItuFaroeseDanish", version=datasets.Version("1.0.0"), description="Faroese-Danish parallel text from ITU Copenhagen"),
    ]

    def _info(self):
        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "origin": datasets.Value("string"),
                    "fo": datasets.Value("string"),
                    "da": datasets.Value("string"),
                }
            ),
            supervised_keys=None,
            homepage="",
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        """Returns SplitGenerators."""
        downloaded_file = dl_manager.download_and_extract(_URL)

        return [
            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_file}),
        ]

    def _generate_examples(self, filepath):
        logger.info("⏳ Generating examples from = %s", filepath)
        texts = {"fao":{}, "dan":{}}
        with open(filepath, encoding="utf-8") as f:
            for line in f:
                line = line.strip()
                if line:
                    code, text = line.split("\t")
                    origin, language = code.split("_")
                    texts[language][origin] = text
        if texts["fao"].keys() != texts["dan"].keys():
            raise ValueError("Faroese and Danish keys don't match")
        guid = 0
        for origin in texts["fao"]:
            yield guid, {
                "id": str(guid),
                "origin": origin,
                "fo": texts["fao"][origin],
                "da": texts["dan"][origin],
            }
            guid += 1