CogComp/mc_taco

1# coding=utf-82# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.3#4# Licensed under the Apache License, Version 2.0 (the "License");5# you may not use this file except in compliance with the License.6# You may obtain a copy of the License at7#8#     http://www.apache.org/licenses/LICENSE-2.09#10# Unless required by applicable law or agreed to in writing, software11# distributed under the License is distributed on an "AS IS" BASIS,12# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.13# See the License for the specific language governing permissions and14# limitations under the License.15"""MC-TACO Dataset."""16 17 18import csv19 20import datasets21 22 23_CITATION = """\24@inproceedings{ZKNR19,25    author = {Ben Zhou, Daniel Khashabi, Qiang Ning and Dan Roth},26    title = {“Going on a vacation” takes longer than “Going for a walk”: A Study of Temporal Commonsense Understanding },27    booktitle = {EMNLP},28    year = {2019},29}30"""31 32_DESCRIPTION = """\33MC-TACO (Multiple Choice TemporAl COmmonsense) is a dataset of 13k question-answer34pairs that require temporal commonsense comprehension. A system receives a sentence35providing context information, a question designed to require temporal commonsense36knowledge, and multiple candidate answers. More than one candidate answer can be plausible.37 38The task is framed as binary classification: givent he context, the question,39and the candidate answer, the task is to determine whether the candidate40answer is plausible ("yes") or not ("no")."""41 42_LICENSE = "Unknown"43 44_URLs = {45    "dev": "https://raw.githubusercontent.com/CogComp/MCTACO/master/dataset/dev_3783.tsv",46    "test": "https://raw.githubusercontent.com/CogComp/MCTACO/master/dataset/test_9442.tsv",47}48 49 50class McTaco(datasets.GeneratorBasedBuilder):51    """MC-TACO Dataset: temporal commonsense knowledge."""52 53    VERSION = datasets.Version("1.1.0")54 55    BUILDER_CONFIGS = [56        datasets.BuilderConfig(57            name="plain_text",58            description="Plain text",59            version=VERSION,60        ),61    ]62 63    def _info(self):64        return datasets.DatasetInfo(65            description=_DESCRIPTION,66            features=datasets.Features(67                {68                    "sentence": datasets.Value("string"),69                    "question": datasets.Value("string"),70                    "answer": datasets.Value("string"),71                    "label": datasets.ClassLabel(names=["no", "yes"]),72                    "category": datasets.ClassLabel(73                        names=["Event Duration", "Event Ordering", "Frequency", "Typical Time", "Stationarity"]74                    ),75                }76            ),77            supervised_keys=None,78            homepage="https://cogcomp.seas.upenn.edu/page/resource_view/125",79            license=_LICENSE,80            citation=_CITATION,81        )82 83    def _split_generators(self, dl_manager):84        """Returns SplitGenerators."""85        data_dir = dl_manager.download_and_extract(_URLs)86        return [87            datasets.SplitGenerator(88                name=datasets.Split.TEST,89                gen_kwargs={90                    "filepath": data_dir["test"],91                },92            ),93            datasets.SplitGenerator(94                name=datasets.Split.VALIDATION,95                gen_kwargs={96                    "filepath": data_dir["dev"],97                },98            ),99        ]100 101    def _generate_examples(self, filepath):102        """Yields examples."""103        with open(filepath, encoding="utf-8") as csv_file:104            csv_reader = csv.reader(105                csv_file,106                delimiter="\t",107            )108            for id_, row in enumerate(csv_reader):109                yield id_, {110                    "sentence": row[0],111                    "question": row[1],112                    "answer": row[2],113                    "label": row[3],114                    "category": row[4],115                }116