chebi_nactem.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.

import os.path
from itertools import count
from pathlib import Path
from typing import Dict, Iterable, List, Tuple

import datasets

from .bigbiohub import kb_features
from .bigbiohub import BigBioConfig
from .bigbiohub import Tasks
from .bigbiohub import parse_brat_file


_LANGUAGES = ["English"]
_PUBMED = True
_LOCAL = False
_CITATION = """\
@inproceedings{Shardlow2018,
  title        = {
    A New Corpus to Support Text Mining for the Curation of Metabolites in the
    {ChEBI} Database
  },
  author       = {
    Shardlow, M J and Nguyen, N and Owen, G and O'Donovan, C and Leach, A and
    McNaught, J and Turner, S and Ananiadou, S
  },
  year         = 2018,
  month        = may,
  booktitle    = {
    Proceedings of the Eleventh International Conference on Language Resources
    and Evaluation ({LREC} 2018)
  },
  location     = {Miyazaki, Japan},
  pages        = {280--285},
  conference   = {
    Eleventh International Conference on Language Resources and Evaluation
    (LREC 2018)
  },
  language     = {en}
}
"""

_DATASETNAME = "chebi_nactem"
_DISPLAYNAME = "CHEBI Corpus"

_DESCRIPTION = """\
The ChEBI corpus contains 199 annotated abstracts and 100 annotated full papers.
All documents in the corpus have been annotated for named entities and relations
between these. In total, our corpus provides over 15000 named entity annotations
and over 6,000 relations between entities.
"""

_HOMEPAGE = "http://www.nactem.ac.uk/chebi"

_LICENSE = "Creative Commons Attribution 4.0 International"

_URLS = {
    _DATASETNAME: "http://www.nactem.ac.uk/chebi/ChEBI.zip",
}

_SUPPORTED_TASKS = [Tasks.NAMED_ENTITY_RECOGNITION, Tasks.RELATION_EXTRACTION]

_SOURCE_VERSION = "1.0.0"

_BIGBIO_VERSION = "1.0.0"


class ChebiNactemDatasset(datasets.GeneratorBasedBuilder):
    """Chemical Entities of Biological Interest (ChEBI) corpus."""

    SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
    BIGBIO_VERSION = datasets.Version(_BIGBIO_VERSION)

    BUILDER_CONFIGS = []
    for subset_id in ["abstr_ann1", "abstr_ann2", "fullpaper"]:
        BUILDER_CONFIGS += [
            BigBioConfig(
                name=f"chebi_nactem_{subset_id}_source",
                version=SOURCE_VERSION,
                description="chebi_nactem source schema",
                schema="source",
                subset_id=f"chebi_nactem_{subset_id}",
            ),
            BigBioConfig(
                name=f"chebi_nactem_{subset_id}_bigbio_kb",
                version=BIGBIO_VERSION,
                description="chebi_nactem BigBio schema",
                schema="bigbio_kb",
                subset_id=f"chebi_nactem_{subset_id}",
            ),
        ]

    DEFAULT_CONFIG_NAME = "chebi_nactem_fullpaper_source"

    def _info(self) -> datasets.DatasetInfo:
        if self.config.schema == "source":
            features = datasets.Features(
                {
                    "document_id": datasets.Value("string"),
                    "text": datasets.Value("string"),
                    "entities": [
                        {
                            "id": datasets.Value("string"),
                            "type": datasets.Value("string"),
                            "text": datasets.Value("string"),
                            "offsets": datasets.Sequence([datasets.Value("int32")]),
                        }
                    ],
                    "relations": [
                        {
                            "id": datasets.Value("string"),
                            "type": datasets.Value("string"),
                            "arg1": datasets.Value("string"),
                            "arg2": datasets.Value("string"),
                        }
                    ],
                }
            )

        elif self.config.schema == "bigbio_kb":
            features = kb_features
        else:
            raise NotImplementedError(self.config.schema)

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

    def _split_generators(self, dl_manager) -> List[datasets.SplitGenerator]:
        """Returns SplitGenerators."""
        urls = _URLS[_DATASETNAME]
        data_dir = dl_manager.download_and_extract(urls)

        subset_paths = {
            "chebi_nactem_abstr_ann1": os.path.join("ChEBI", "abstracts", "Annotator1"),
            "chebi_nactem_abstr_ann2": os.path.join("ChEBI", "abstracts", "Annotator2"),
            "chebi_nactem_fullpaper": os.path.join("ChEBI", "fullpapers"),
        }
        subset_dir = os.path.join(data_dir, subset_paths[self.config.subset_id])
        return [
            datasets.SplitGenerator(
                name=datasets.Split.TRAIN,
                gen_kwargs={"file_paths": dl_manager.iter_files(subset_dir)},
            )
        ]

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

        uid = count(0)

        for idx, file_path in enumerate(file_paths):
            if os.path.basename(file_path).endswith(".txt"):
                contents = parse_brat_file(
                    Path(file_path), annotation_file_suffixes=[".ann"]
                )

                if self.config.schema == "source":
                    yield idx, {
                        "document_id": contents["document_id"],
                        "text": contents["text"],
                        "entities": contents["text_bound_annotations"],
                        "relations": [
                            {
                                "id": relation["id"],
                                "type": relation["type"],
                                "arg1": relation["head"]["ref_id"],
                                "arg2": relation["tail"]["ref_id"],
                            }
                            for relation in contents["relations"]
                        ],
                    }

                elif self.config.schema == "bigbio_kb":
                    yield idx, {
                        "id": next(uid),
                        "document_id": contents["document_id"],
                        "passages": [
                            {
                                "id": next(uid),
                                "type": "",
                                "text": [contents["text"]],
                                "offsets": [(0, len(contents["text"]))],
                            }
                        ],
                        "entities": [
                            {
                                "id": f"{idx}_{entity['id']}",
                                "type": entity["type"],
                                "offsets": entity["offsets"],
                                "text": entity["text"],
                                "normalized": [],
                            }
                            for entity in contents["text_bound_annotations"]
                        ],
                        "events": [],
                        "coreferences": [],
                        "relations": [
                            {
                                "id": f"{idx}_{relation['id']}",
                                "type": relation["type"],
                                "arg1_id": f"{idx}_{relation['head']['ref_id']}",
                                "arg2_id": f"{idx}_{relation['tail']['ref_id']}",
                                "normalized": [],
                            }
                            for relation in contents["relations"]
                        ],
                    }
                else:
                    raise NotImplementedError(self.config.schema)