wsd_polish_datasets.py

# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# 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.
import itertools
import json
from typing import Sequence

import datasets

logger = datasets.logging.get_logger(__name__)

_CITATION = """\
@InProceedings{10.1007/978-3-031-08754-7_70,
  author="Janz, Arkadiusz
  and Dziob, Agnieszka
  and Oleksy, Marcin
  and Baran, Joanna",
  editor="Groen, Derek
  and de Mulatier, Cl{\'e}lia
  and Paszynski, Maciej
  and Krzhizhanovskaya, Valeria V.
  and Dongarra, Jack J.
  and Sloot, Peter M. A.",
  title="A Unified Sense Inventory for Word Sense Disambiguation in Polish",
  booktitle="Computational Science -- ICCS 2022",
  year="2022",
  publisher="Springer International Publishing",
  address="Cham",
  pages="682--689",
  isbn="978-3-031-08754-7"
}
"""
_DESCRIPTION = """\
Polish WSD training data manually annotated by experts according to plWordNet-4.2.
"""

_LICENSE = "cc-by-4.0"

_BASE_URL = "https://huggingface.co/datasets/clarin-knext/wsd_polish_datasets/resolve/main/data/"

_CORPUS_NAMES = [
    "sherlock",
    "skladnica",
    "wikiglex",
    "emoglex",
    "walenty",
    "kpwr",
    "kpwr-100",
]

_DATA_TYPES = [
    "sentence",
    "text",
]

_URLS = {
    "text": {corpus: f"{_BASE_URL}{corpus}_text.jsonl" for corpus in _CORPUS_NAMES},
    "sentence": {
        corpus: f"{_BASE_URL}{corpus}_sentences.jsonl" for corpus in _CORPUS_NAMES
    },
}


class WsdPolishBuilderConfig(datasets.BuilderConfig):
    def __init__(
        self,
        data_urls: Sequence[str],
        corpus: str,
        data_type: str,
        **kwargs,
    ):
        super(WsdPolishBuilderConfig, self).__init__(
            name=f"{corpus}_{data_type}",
            version=datasets.Version("1.0.0"),
            **kwargs,
        )

        self.data_type = data_type
        self.corpus = corpus
        self.data_urls = data_urls
        if self.data_type not in _DATA_TYPES:
            raise ValueError(
                f"Corpus type {self.data_type} is not supported. Enter one of: {_DATA_TYPES}"
            )
        if self.corpus not in (*_CORPUS_NAMES, "all"):
            raise ValueError(
                f"Corpus name `{self.corpus}` is not available. Enter one of: {(*_CORPUS_NAMES, 'all')}"
            )


class WsdPolishDataset(datasets.GeneratorBasedBuilder):
    """Polish WSD training data"""

    BUILDER_CONFIGS = [
        WsdPolishBuilderConfig(
            corpus=corpus_name,
            data_type=data_type,
            data_urls=[_URLS[data_type][corpus_name]],
            description=f"Data part covering `{corpus_name}` corpora in `{data_type}` segmentation.",
        )
        for corpus_name, data_type in itertools.product(_CORPUS_NAMES, _DATA_TYPES)
    ]
    BUILDER_CONFIGS.extend(
        [
            WsdPolishBuilderConfig(
                corpus="all",
                data_type=data_type,
                data_urls=list(_URLS[data_type].values()),
                description=f"Data part covering `all` corpora in `{data_type}` segmentation.",
            )
            for data_type in _DATA_TYPES
        ]
    )

    DEFAULT_CONFIG_NAME = "all_text"

    def _info(self) -> datasets.DatasetInfo:
        text_features = {
            "text": datasets.Value("string"),
            "tokens": datasets.features.Sequence(
                dict(
                    {
                        "position": datasets.features.Sequence(
                            length=2,
                            feature=datasets.Value("int32"),
                        ),
                        "orth": datasets.Value("string"),
                        "lemma": datasets.Value("string"),
                        "pos": datasets.Value("string"),
                    }
                ),
            ),
            "phrases": datasets.features.Sequence(
                dict(
                    {
                        "indices": datasets.features.Sequence(
                            feature=datasets.Value("int32")
                        ),
                        "head": datasets.Value("int32"),
                        "lemma": datasets.Value("string"),
                    }
                ),
            ),
            "wsd": datasets.features.Sequence(
                dict(
                    {
                        "index": datasets.Value("int32"),
                        "plWN_syn_id": datasets.Value("string"),
                        "plWN_lex_id": datasets.Value("string"),
                        "PWN_syn_id": datasets.Value("string"),
                        "bn_syn_id": datasets.Value("string"),
                        "mapping_relation": datasets.Value("string")
                    }
                ),
            ),
        }
        if self.config.data_type == "sentence":
            features = datasets.Features(
                {
                    "sentences": datasets.features.Sequence(text_features),
                }
            )
        else:
            features = datasets.Features(text_features)

        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            supervised_keys=None,
            license=_LICENSE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager):
        filepaths = dl_manager.download_and_extract(self.config.data_urls)
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={
                    "filepaths": filepaths,
                },
            ),
        ]

    def _generate_examples(self, filepaths: Sequence[str]):
        key_iter = 0
        for filepath in filepaths:
            with open(filepath, encoding="utf-8") as f:
                for data in (json.loads(line) for line in f):
                    if self.config.data_type == "sentence":
                        yield key_iter, {
                            "sentences": [
                                self._process_example(sent)
                                for sent in data["sentences"]
                            ]
                        }
                    else:
                        data.pop("context_file")
                        yield key_iter, self._process_example(data)

                    key_iter += 1

    @staticmethod
    def _process_example(data: dict) -> dict:
        return {
            "text": data["text"],
            "tokens": [
                {
                    "position": tok["position"],
                    "orth": tok["orth"],
                    "lemma": tok["lemma"],
                    "pos":tok["pos"],
                }
                for tok in data["tokens"]
            ],
            "wsd": [
                {
                    "index": tok["index"],
                    "plWN_syn_id": tok["plWN_syn_id"],
                    "plWN_lex_id": tok["plWN_lex_id"],
                    # "PWN_syn_id": tok["PWN_syn_id"],
                    # "bn_syn_id": tok["bn_syn_id"],
                    # "mapping_relation":  tok["mapping_relation"],

                }
                for tok in data["wsd"]
            ],
            "phrases": [
                {
                    "indices": tok["indices"],
                    "head": tok["head"],
                    "lemma": tok["lemma"],
                }
                for tok in data["phrases"]
            ],
        }