Datasets:
Sub-tasks:
natural-language-inference
Languages:
Catalan
Multilinguality:
monolingual
Annotations Creators:
professional translators
License:
| # Loading script for the COPA-ca dataset. | |
| import json | |
| import datasets | |
| logger = datasets.logging.get_logger(__name__) | |
| _CITATION = "" | |
| _DESCRIPTION = """\ | |
| The COPA-ca dataset (Choice of plausible alternatives in Catalan) is a professional translation of the English COPA dataset into Catalan, commissioned by BSC LangTech Unit. The dataset consists of 1000 premises, each given a question and two choices with a label encoding which of the choices is more plausible given the annotator. | |
| The dataset is split into 400 training samples, 100 validation samples, and 500 test samples. It includes the following features: 'premise', 'choice1', 'choice2', 'label', 'question', 'changed' (boolean). | |
| This work is licensed under a Attribution-ShareAlike 4.0 International License. | |
| """ | |
| _HOMEPAGE = "https://zenodo.org/record/8124398" | |
| _URL = "https://huggingface.co/datasets/projecte-aina/copa-ca/resolve/main/" | |
| _TRAIN_FILE = "copa-ca.train.jsonl" | |
| _DEV_FILE = "copa-ca.val.jsonl" | |
| _TEST_FILE = "copa-ca.test.jsonl" | |
| class copaCaConfig(datasets.BuilderConfig): | |
| """ Builder config for the COPA-ca dataset """ | |
| def __init__(self, **kwargs): | |
| """BuilderConfig for COPA-ca. | |
| Args: | |
| **kwargs: keyword arguments forwarded to super. | |
| """ | |
| super(copaCaConfig, self).__init__(**kwargs) | |
| class copaCa(datasets.GeneratorBasedBuilder): | |
| """ COPA-ca Dataset """ | |
| BUILDER_CONFIGS = [ | |
| copaCaConfig( | |
| name="copa-ca", | |
| version=datasets.Version("1.0.1"), | |
| description="COPA-ca dataset", | |
| ), | |
| ] | |
| def _info(self): | |
| return datasets.DatasetInfo( | |
| description=_DESCRIPTION, | |
| features=datasets.Features( | |
| { | |
| "premise": datasets.Value("string"), | |
| "choice1": datasets.Value("string"), | |
| "choice2": datasets.Value("string"), | |
| "question": datasets.Value("string"), | |
| 'label': datasets.features.ClassLabel(names=['1', '2']), | |
| "idx": datasets.Value("int64"), | |
| "changed": datasets.Value("bool"), | |
| } | |
| ), | |
| homepage=_HOMEPAGE, | |
| citation=_CITATION, | |
| ) | |
| def _split_generators(self, dl_manager): | |
| """Returns SplitGenerators.""" | |
| urls_to_download = { | |
| "train": f"{_URL}{_TRAIN_FILE}", | |
| "dev": f"{_URL}{_DEV_FILE}", | |
| "test": f"{_URL}{_TEST_FILE}", | |
| } | |
| downloaded_files = dl_manager.download_and_extract(urls_to_download) | |
| return [ | |
| datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files["train"]}), | |
| datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files["dev"]}), | |
| datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files["test"]}), | |
| ] | |
| def _generate_examples(self, filepath): | |
| with open(filepath, encoding='utf-8') as f: | |
| for i, line in enumerate(f): | |
| data = json.loads(line) | |
| yield i, { | |
| 'premise': data['premise'], | |
| 'choice1': data['choice1'], | |
| 'choice2': data['choice2'], | |
| 'question': data['question'], | |
| 'label': str(data['label']), | |
| 'idx': data['idx'], | |
| 'changed': data['changed'] | |
| } | |