# coding=utf-8 # Copyright 2022 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. """The MultiTACRED Relation Classification dataset in various languages""" import itertools import json import os import datasets _CITATION = """\ @inproceedings{hennig-etal-2023-multitacred, title = "MultiTACRED: A Multilingual Version of the TAC Relation Extraction Dataset", author = "Hennig, Leonhard and Thomas, Philippe and Möller, Sebastian", booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)", month = jul, year = "2023", address = "Online and Toronto, Canada", publisher = "Association for Computational Linguistics", } @inproceedings{zhang-etal-2017-position, title = "Position-aware Attention and Supervised Data Improve Slot Filling", author = "Zhang, Yuhao and Zhong, Victor and Chen, Danqi and Angeli, Gabor and Manning, Christopher D.", booktitle = "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing", month = sep, year = "2017", address = "Copenhagen, Denmark", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/D17-1004", doi = "10.18653/v1/D17-1004", pages = "35--45", } @inproceedings{alt-etal-2020-tacred, title = "{TACRED} Revisited: A Thorough Evaluation of the {TACRED} Relation Extraction Task", author = "Alt, Christoph and Gabryszak, Aleksandra and Hennig, Leonhard", booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics", month = jul, year = "2020", address = "Online", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/2020.acl-main.142", doi = "10.18653/v1/2020.acl-main.142", pages = "1558--1569", } @inproceedings{DBLP:conf/aaai/StoicaPP21, author = {George Stoica and Emmanouil Antonios Platanios and Barnab{\'{a}}s P{\'{o}}czos}, title = {Re-TACRED: Addressing Shortcomings of the {TACRED} Dataset}, booktitle = {Thirty-Fifth {AAAI} Conference on Artificial Intelligence, {AAAI} 2021, Thirty-Third Conference on Innovative Applications of Artificial Intelligence, {IAAI} 2021, The Eleventh Symposium on Educational Advances in Artificial Intelligence, {EAAI} 2021, Virtual Event, February 2-9, 2021}, pages = {13843--13850}, publisher = {{AAAI} Press}, year = {2021}, url = {https://ojs.aaai.org/index.php/AAAI/article/view/17631}, } """ _DESCRIPTION = """\ MultiTACRED is a multilingual version of the large-scale TAC Relation Extraction Dataset (https://nlp.stanford.edu/projects/tacred). It covers 12 typologically diverse languages from 9 language families, and was created by the Speech & Language Technology group of DFKI (https://www.dfki.de/slt) by machine-translating the instances of the original TACRED dataset and automatically projecting their entity annotations. For details of the original TACRED's data collection and annotation process, see the Stanford paper (https://aclanthology.org/D17-1004/). Translations are syntactically validated by checking the correctness of the XML tag markup. Any translations with an invalid tag structure, e.g. missing or invalid head or tail tag pairs, are discarded (on average, 2.3% of the instances). Languages covered are: Arabic, Chinese, Finnish, French, German, Hindi, Hungarian, Japanese, Polish, Russian, Spanish, Turkish. Intended use is supervised relation classification. Audience - researchers. Please see our ACL paper (https://arxiv.org/abs/2305.04582) for full details. NOTE: This Datasetreader supports a reduced version of the original TACRED JSON format with the following changes: - Removed fields: stanford_pos, stanford_ner, stanford_head, stanford_deprel, docid The motivation for this is that we want to support additional languages, for which these fields were not required or available. The reader expects the specification of a language-specific configuration specifying the variant (original, revisited or retacred) and the language (as a two-letter iso code). The DatasetReader changes the offsets of the following fields, to conform with standard Python usage (see _generate_examples()): - subj_end to subj_end + 1 (make end offset exclusive) - obj_end to obj_end + 1 (make end offset exclusive) NOTE 2: The MultiTACRED dataset offers an additional 'split', namely the backtranslated test data (translated to a target language and then back to English). To access this split, use dataset['backtranslated_test']. You can find the TACRED dataset reader for the English version of the dataset at https://huggingface.co/datasets/DFKI-SLT/tacred. """ _HOMEPAGE = "https://github.com/DFKI-NLP/MultiTACRED" _LICENSE = "LDC" _URL = "https://catalog.ldc.upenn.edu/LDC2023_TODO" # The HuggingFace dataset library don't host the datasets but only point to the original files # This can be an arbitrary nested dict/list of URLs (see below in `_split_generators` method) _PATCH_URLs = { "dev": "https://raw.githubusercontent.com/DFKI-NLP/tacrev/master/patch/dev_patch.json", "test": "https://raw.githubusercontent.com/DFKI-NLP/tacrev/master/patch/test_patch.json", } _RETACRED_PATCH_URLs = { "train": "https://raw.githubusercontent.com/gstoica27/Re-TACRED/master/Re-TACRED/train_id2label.json", "dev": "https://raw.githubusercontent.com/gstoica27/Re-TACRED/master/Re-TACRED/dev_id2label.json", "test": "https://raw.githubusercontent.com/gstoica27/Re-TACRED/master/Re-TACRED/test_id2label.json" } _BACKTRANSLATION_TEST_SPLIT = "backtranslated_test" _RETACRED = "retacred" _REVISITED = "revisited" _ORIGINAL = "original" _VERSION = datasets.Version("1.1.0") _LANGS = [ "ar", "de", "es", "fi", "fr", "hi", "hu", "ja", "pl", "ru", "tr", "zh", ] _CLASS_LABELS = [ "no_relation", "org:alternate_names", "org:city_of_headquarters", "org:country_of_headquarters", "org:dissolved", "org:founded", "org:founded_by", "org:member_of", "org:members", "org:number_of_employees/members", "org:parents", "org:political/religious_affiliation", "org:shareholders", "org:stateorprovince_of_headquarters", "org:subsidiaries", "org:top_members/employees", "org:website", "per:age", "per:alternate_names", "per:cause_of_death", "per:charges", "per:children", "per:cities_of_residence", "per:city_of_birth", "per:city_of_death", "per:countries_of_residence", "per:country_of_birth", "per:country_of_death", "per:date_of_birth", "per:date_of_death", "per:employee_of", "per:origin", "per:other_family", "per:parents", "per:religion", "per:schools_attended", "per:siblings", "per:spouse", "per:stateorprovince_of_birth", "per:stateorprovince_of_death", "per:stateorprovinces_of_residence", "per:title", ] _RETACRED_CLASS_LABELS = [ "no_relation", "org:alternate_names", "org:city_of_branch", "org:country_of_branch", "org:dissolved", "org:founded", "org:founded_by", "org:member_of", "org:members", "org:number_of_employees/members", "org:political/religious_affiliation", "org:shareholders", "org:stateorprovince_of_branch", "org:top_members/employees", "org:website", "per:age", "per:cause_of_death", "per:charges", "per:children", "per:cities_of_residence", "per:city_of_birth", "per:city_of_death", "per:countries_of_residence", "per:country_of_birth", "per:country_of_death", "per:date_of_birth", "per:date_of_death", "per:employee_of", "per:identity", "per:origin", "per:other_family", "per:parents", "per:religion", "per:schools_attended", "per:siblings", "per:spouse", "per:stateorprovince_of_birth", "per:stateorprovince_of_death", "per:stateorprovinces_of_residence", "per:title" ] _NER_CLASS_LABELS = [ "LOCATION", "ORGANIZATION", "PERSON", "DATE", "MONEY", "PERCENT", "TIME", "CAUSE_OF_DEATH", "CITY", "COUNTRY", "CRIMINAL_CHARGE", "EMAIL", "HANDLE", "IDEOLOGY", "NATIONALITY", "RELIGION", "STATE_OR_PROVINCE", "TITLE", "URL", "NUMBER", "ORDINAL", "MISC", "DURATION", "O", ] _DESC_TEXTS = {_ORIGINAL: 'The original TACRED.', _REVISITED: 'TACRED Revisited (corrected labels for 5k most challenging examples in dev and test split).', _RETACRED: 'Relabeled TACRED (corrected labels for all splits and pruned)'} def convert_ptb_token(token: str) -> str: """Convert PTB tokens to normal tokens""" return { "-lrb-": "(", "-rrb-": ")", "-lsb-": "[", "-rsb-": "]", "-lcb-": "{", "-rcb-": "}", }.get(token.lower(), token) class MultiTacredConfig(datasets.BuilderConfig): """BuilderConfig for MultiTacred.""" def __init__(self, label_variant, language, **kwargs): """BuilderConfig for MultiTacred. Args: label_variant: `string`, source of labels, i.e. 'original', 'revisited' or 'retacred' language: `string`, 2-letter ISO 639-1 language code **kwargs: keyword arguments forwarded to super. """ super(MultiTacredConfig, self).__init__(version=_VERSION, **kwargs) self.language = language self.label_variant = label_variant class MultiTacred(datasets.GeneratorBasedBuilder): """MultiTACRED is a multilingual version of the large-scale TAC Relation Extraction Dataset (LDC2018T24).""" BUILDER_CONFIGS = [ MultiTacredConfig( name=f"{label_variant}-{language}", language=language, label_variant=label_variant, description=f"{_DESC_TEXTS[label_variant]} examples in language '{language}'.", ) for (language, label_variant) in itertools.product(_LANGS, [_ORIGINAL, _REVISITED, _RETACRED]) ] @property def manual_download_instructions(self): return ( "To use MultiTACRED you have to download it manually. " "It is available via the LDC at https://catalog.ldc.upenn.edu/LDC2023_TODO" "Please extract all files in one folder and load the a language with: " "`datasets.load_dataset('DFKI-SLT/multitacred', name='variant-language', data_dir='path/to/folder/folder_name')`." ) def _info(self): features = datasets.Features( { "id": datasets.Value("string"), "token": datasets.Sequence(datasets.Value("string")), "subj_start": datasets.Value("int32"), "subj_end": datasets.Value("int32"), "subj_type": datasets.ClassLabel(names=_NER_CLASS_LABELS), "obj_start": datasets.Value("int32"), "obj_end": datasets.Value("int32"), "obj_type": datasets.ClassLabel(names=_NER_CLASS_LABELS), "relation": datasets.ClassLabel( names=_RETACRED_CLASS_LABELS if self.config.label_variant == _RETACRED else _CLASS_LABELS), } ) return datasets.DatasetInfo( # This is the description that will appear on the datasets page. description=_DESCRIPTION, # This defines the different columns of the dataset and their types features=features, # Here we define them above because they are different between the two configurations # If there's a common (input, target) tuple from the features, # specify them here. They'll be used if as_supervised=True in # builder.as_dataset. supervised_keys=None, # Homepage of the dataset for documentation homepage=_HOMEPAGE, # License for the dataset if available license=_LICENSE, # Citation for the dataset citation=_CITATION, ) def _split_generators(self, dl_manager): """Returns SplitGenerators.""" # If several configurations are possible (listed in BUILDER_CONFIGS), the configuration selected by the user is in self.config.name # dl_manager is a datasets.download.DownloadManager that can be used to download and extract URLs # It can accept any type or nested list/dict and will give back the same structure with the url replaced with path to local files. # By default the archives will be extracted and a path to a cached folder where they are extracted is returned instead of the archive data_dir = os.path.abspath(os.path.expanduser(dl_manager.manual_dir)) if not os.path.exists(data_dir): raise FileNotFoundError( "{} does not exist. Make sure you insert a manual dir via `datasets.load_dataset(" "'DFKI-SLT/multitacred', name=..., data_dir=...)` that includes the unzipped files from the " "MULTITACRED_LDC zip. Manual download instructions: {}".format( data_dir, self.manual_download_instructions ) ) patch_files = {} if self.config.label_variant == _REVISITED: patch_files = dl_manager.download_and_extract(_PATCH_URLs) elif self.config.label_variant == _RETACRED: patch_files = dl_manager.download_and_extract(_RETACRED_PATCH_URLs) return [ datasets.SplitGenerator( name=datasets.Split.TRAIN, gen_kwargs={ "filepath": os.path.join(data_dir, self.config.language, f"train_{self.config.language}.json"), "patch_filepath": patch_files.get("train"), }, ), datasets.SplitGenerator( name=datasets.Split.TEST, gen_kwargs={ "filepath": os.path.join(data_dir, self.config.language, f"test_{self.config.language}.json"), "patch_filepath": patch_files.get("test"), }, ), datasets.SplitGenerator( name=datasets.Split.VALIDATION, gen_kwargs={ "filepath": os.path.join(data_dir, self.config.language, f"dev_{self.config.language}.json"), "patch_filepath": patch_files.get("dev"), }, ), datasets.SplitGenerator( name=_BACKTRANSLATION_TEST_SPLIT, gen_kwargs={ "filepath": os.path.join(data_dir, self.config.language, f"test_en_{self.config.language}_bt.json"), "patch_filepath": patch_files.get("test"), }, ), ] def _generate_examples(self, filepath, patch_filepath): """Yields examples.""" # This method will receive as arguments the `gen_kwargs` defined in the previous `_split_generators` method. # It is in charge of opening the given file and yielding (key, example) tuples from the dataset # The key is not important, it's more here for legacy reason (legacy from tfds) patch_examples = {} if patch_filepath is not None: with open(patch_filepath, encoding="utf-8") as f: if self.config.label_variant == _REVISITED: patch_examples = {example["id"]: example for example in json.load(f)} elif self.config.label_variant == _RETACRED: patch_examples = {_id: {"id": _id, "relation": label} for _id, label in json.load(f).items()} with open(filepath, encoding="utf-8") as f: data = json.load(f) for example in data: id_ = example["id"] if id_ in patch_examples: example.update(patch_examples[id_]) elif self.config.label_variant == _RETACRED: # RE-TACRED was pruned, skip example if its id is not in patch_examples continue yield id_, { "id": example["id"], "token": [convert_ptb_token(token) for token in example["token"]], "subj_start": example["subj_start"], "subj_end": example["subj_end"] + 1, # make end offset exclusive "subj_type": example["subj_type"], "obj_start": example["obj_start"], "obj_end": example["obj_end"] + 1, # make end offset exclusive "obj_type": example["obj_type"], "relation": example["relation"], }