# coding=utf-8 # Copyright 2020 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. """Children's Book Test Dataset.""" import datasets _CITATION = """\ @misc{hill2016goldilocks, title={The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations}, author={Felix Hill and Antoine Bordes and Sumit Chopra and Jason Weston}, year={2016}, eprint={1511.02301}, archivePrefix={arXiv}, primaryClass={cs.CL} } """ _DESCRIPTION = """\ The Children’s Book Test (CBT) is designed to measure directly how well language models can exploit wider linguistic context. The CBT is built from books that are freely available. """ _HOMEPAGE = "https://research.fb.com/downloads/babi/" _LICENSE = """GNU Free Documentation License v1.3""" ZIP_URL = "http://www.thespermwhale.com/jaseweston/babi/CBTest.tgz" dir = "CBTest/data/" paths = { "raw": {"train": dir + "cbt_train.txt", "valid": dir + "cbt_valid.txt", "test": dir + "cbt_test.txt"}, "V": { "train": dir + "cbtest_V_train.txt", "valid": dir + "cbtest_V_valid_2000ex.txt", "test": dir + "cbtest_V_test_2500ex.txt", }, "P": { "train": dir + "cbtest_P_train.txt", "valid": dir + "cbtest_P_valid_2000ex.txt", "test": dir + "cbtest_P_test_2500ex.txt", }, "NE": { "train": dir + "cbtest_NE_train.txt", "valid": dir + "cbtest_NE_valid_2000ex.txt", "test": dir + "cbtest_NE_test_2500ex.txt", }, "CN": { "train": dir + "cbtest_CN_train.txt", "valid": dir + "cbtest_CN_valid_2000ex.txt", "test": dir + "cbtest_CN_test_2500ex.txt", }, } class Cbt(datasets.GeneratorBasedBuilder): """TODO: Short description of my dataset.""" VERSION = datasets.Version("1.1.0") BUILDER_CONFIGS = [ datasets.BuilderConfig( name="raw", version=VERSION, description="This part of my dataset covers the raw CBT books" ), datasets.BuilderConfig( name="V", version=VERSION, description="This part of my dataset covers the verb answer CBT dataset" ), datasets.BuilderConfig( name="P", version=VERSION, description="This part of my dataset covers the preposition answer CBT dataset" ), datasets.BuilderConfig( name="NE", version=VERSION, description="This part of my dataset covers the named entity answer CBT dataset", ), datasets.BuilderConfig( name="CN", version=VERSION, description="This part of my dataset covers the common noun answer CBT dataset" ), ] def _info(self): if self.config.name in ["V", "P", "NE", "CN"]: features = datasets.Features( { "sentences": datasets.Sequence(datasets.Value("string")), # There are 20 sentences "question": datasets.Value("string"), "answer": datasets.Value("string"), "options": datasets.Sequence(datasets.Value("string")), } ) else: # This is an example to show how to have different features for "first_domain" and "second_domain" features = datasets.Features({"title": datasets.Value("string"), "content": datasets.Value("string")}) 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.""" my_urls = ZIP_URL # Cannot download just one single type as it is a compressed file. archive = dl_manager.download(my_urls) return [ datasets.SplitGenerator( name=datasets.Split.TRAIN, # These kwargs will be passed to _generate_examples gen_kwargs={"filepath": paths[self.config.name]["train"], "files": dl_manager.iter_archive(archive)}, ), datasets.SplitGenerator( name=datasets.Split.TEST, # These kwargs will be passed to _generate_examples gen_kwargs={"filepath": paths[self.config.name]["test"], "files": dl_manager.iter_archive(archive)}, ), datasets.SplitGenerator( name=datasets.Split.VALIDATION, # These kwargs will be passed to _generate_examples gen_kwargs={"filepath": paths[self.config.name]["valid"], "files": dl_manager.iter_archive(archive)}, ), ] def _generate_examples(self, filepath, files): """Yields examples as (key, example) tuples.""" for path, f in files: if path == filepath: if self.config.name != "raw": sentences = [] example_idx = 0 for idx, line in enumerate(f): line = line.decode("utf-8") if line.strip() == "": continue elif line.split()[0] == "21": splitline = line.split("\t") # question, answer options are tab separated question = splitline[0] answer = splitline[1] options = splitline[-1] question = question[2:].strip() # The first two indices contain `21`. answer = answer.strip() options = options.strip().split("|") yield example_idx, { "sentences": sentences, "question": question, "options": options, "answer": answer, } sentences = [] example_idx += 1 else: if len(line.split()[0]) == 1: sentences.append(line[1:].strip()) else: sentences.append(line[2:].strip()) # Text might contain double spaces. else: book_idx = 0 book_sentences = [] for idx, line in enumerate(f): line = line.decode("utf-8") if line[:12] == "_BOOK_TITLE_": if idx == 0: # First line: title = line.split(":")[1].strip() else: yield book_idx, { "title": title, "content": "".join(book_sentences), } title = line.split(":")[1].strip() book_sentences = [] book_idx += 1 else: book_sentences.append(line) else: yield book_idx, { "title": title, "content": "".join(book_sentences), } book_sentences = [] book_idx += 1 break