etos.py

# coding=utf-8
# Copyright 2022 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.

from pathlib import Path
from typing import Dict, List, Tuple

import conllu
import datasets

from seacrowd.utils import schemas
from seacrowd.utils.configs import SEACrowdConfig
from seacrowd.utils.constants import Licenses, Tasks

_CITATION = """\
@INPROCEEDINGS{10053062,
  author={Samsuri, Mukhlizar Nirwan and Yuliawati, Arlisa and Alfina, Ika},
  booktitle={2022 5th International Seminar on Research of Information Technology and Intelligent Systems (ISRITI)},
  title={A Comparison of Distributed, PAM, and Trie Data Structure Dictionaries in Automatic Spelling Correction for Indonesian Formal Text},
  year={2022},
  pages={525-530},
  keywords={Seminars;Dictionaries;Data structures;Intelligent systems;Information technology;automatic spelling correction;distributed dictionary;non-word error;trie data structure;Partition Around Medoids},
  doi={10.1109/ISRITI56927.2022.10053062},
  url = {https://ieeexplore.ieee.org/document/10053062},
}
"""

_DATASETNAME = "etos"

_DESCRIPTION = """\
ETOS (Ejaan oTOmatiS) is a dataset for parts-of-speech (POS) tagging for formal Indonesian
text. It consists of 200 sentences, with 4,323 tokens in total, annotated following the
CoNLL format.
"""

_HOMEPAGE = "https://github.com/ir-nlp-csui/etos"

_LANGUAGES = ["ind"]

_LICENSE = Licenses.AGPL_3_0.value

_LOCAL = False

_URLS = "https://raw.githubusercontent.com/ir-nlp-csui/etos/main/gold_standard.conllu"

_SUPPORTED_TASKS = [Tasks.POS_TAGGING]

_SOURCE_VERSION = "1.0.0"

_SEACROWD_VERSION = "2024.06.20"


class ETOSDataset(datasets.GeneratorBasedBuilder):
    """
    ETOS is an Indonesian parts-of-speech (POS) tagging dataset from https://github.com/ir-nlp-csui/etos.
    """

    SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
    SEACROWD_VERSION = datasets.Version(_SEACROWD_VERSION)

    UPOS_TAGS = [
        "NOUN",
        "PUNCT",
        "ADP",
        "NUM",
        "SYM",
        "SCONJ",
        "ADJ",
        "PART",
        "DET",
        "CCONJ",
        "PROPN",
        "PRON",
        "X",
        "_",
        "ADV",
        "INTJ",
        "VERB",
        "AUX",
    ]

    BUILDER_CONFIGS = [
        SEACrowdConfig(
            name=f"{_DATASETNAME}_source",
            version=datasets.Version(_SOURCE_VERSION),
            description=f"{_DATASETNAME} source schema",
            schema="source",
            subset_id=f"{_DATASETNAME}",
        ),
        SEACrowdConfig(
            name=f"{_DATASETNAME}_seacrowd_seq_label",
            version=datasets.Version(_SOURCE_VERSION),
            description=f"{_DATASETNAME} sequence labeling schema",
            schema="seacrowd_seq_label",
            subset_id=f"{_DATASETNAME}",
        ),
    ]

    DEFAULT_CONFIG_NAME = f"{_DATASETNAME}_source"

    def _info(self) -> datasets.DatasetInfo:
        if self.config.schema == "source":
            features = datasets.Features(
                {
                    "sent_id": datasets.Value("string"),
                    "text": datasets.Value("string"),
                    "tokens": datasets.Sequence(datasets.Value("string")),
                    "lemmas": datasets.Sequence(datasets.Value("string")),
                    "upos": datasets.Sequence(datasets.features.ClassLabel(names=self.UPOS_TAGS)),
                    "xpos": datasets.Sequence(datasets.Value("string")),
                    "feats": datasets.Sequence(datasets.Value("string")),
                    "head": datasets.Sequence(datasets.Value("string")),
                    "deprel": datasets.Sequence(datasets.Value("string")),
                    "deps": datasets.Sequence(datasets.Value("string")),
                    "misc": datasets.Sequence(datasets.Value("string")),
                }
            )

        elif self.config.schema == "seacrowd_seq_label":
            features = schemas.seq_label_features(self.UPOS_TAGS)

        else:
            raise ValueError(f"Invalid schema: '{self.config.schema}'")

        return datasets.DatasetInfo(
            description=_DESCRIPTION,
            features=features,
            homepage=_HOMEPAGE,
            license=_LICENSE,
            citation=_CITATION,
        )

    def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
        """
        Returns SplitGenerators.
        """

        train_path = dl_manager.download_and_extract(_URLS)

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

    def _generate_examples(self, filepath: Path, split: str) -> Tuple[int, Dict]:
        """
        Yields examples as (key, example) tuples.
        """

        with open(filepath, "r", encoding="utf-8") as data_file:
            tokenlist = list(conllu.parse_incr(data_file))

        for idx, sent in enumerate(tokenlist):
            if "sent_id" in sent.metadata:
                sent_id = sent.metadata["sent_id"]
            else:
                sent_id = idx

            tokens = [token["form"] for token in sent]

            if "text" in sent.metadata:
                txt = sent.metadata["text"]
            else:
                txt = " ".join(tokens)

            if self.config.schema == "source":
                yield idx, {
                    "sent_id": str(sent_id),
                    "text": txt,
                    "tokens": tokens,
                    "lemmas": [token["lemma"] for token in sent],
                    "upos": [token["upos"] for token in sent],
                    "xpos": [token["xpos"] for token in sent],
                    "feats": [str(token["feats"]) for token in sent],
                    "head": [str(token["head"]) for token in sent],
                    "deprel": [str(token["deprel"]) for token in sent],
                    "deps": [str(token["deps"]) for token in sent],
                    "misc": [str(token["misc"]) for token in sent],
                }

            elif self.config.schema == "seacrowd_seq_label":
                yield idx, {
                    "id": str(sent_id),
                    "tokens": tokens,
                    "labels": [token["upos"] for token in sent],
                }

            else:
                raise ValueError(f"Invalid schema: '{self.config.schema}'")