import json import math import datasets logger = datasets.logging.get_logger(__name__) _CITATION = """\ @ONLINE {wikidump, author = {Wikimedia Foundation}, title = {Wikimedia Downloads}, url = {https://dumps.wikimedia.org} } """ _DESCRIPTION = """\ Wikipedia version split into plain text snippets for dense semantic indexing. """ _LICENSE = ( "This work is licensed under the Creative Commons Attribution-ShareAlike " "3.0 Unported License. To view a copy of this license, visit " "http://creativecommons.org/licenses/by-sa/3.0/ or send a letter to " "Creative Commons, PO Box 1866, Mountain View, CA 94042, USA." ) def wiki40b_article_snippets(article, passage_len=100, overlap=0): paragraphs = article["text"].split("\n") aticle_idx = paragraphs.index("_START_ARTICLE_") + 1 article_title = paragraphs[aticle_idx] if aticle_idx < len(paragraphs) else "" section_indices = [i + 1 for i, par in enumerate(paragraphs[:-1]) if par == "_START_SECTION_"] par_tabs = [par.split(" ") for par in paragraphs] word_map = [ (i, len(" ".join(par[:j])), w) for i, par in enumerate(par_tabs) if not par[0].startswith("_START_") for j, w in enumerate(par) if i > 0 ] step_size = passage_len - overlap passages = [] for i in range(math.ceil(len(word_map) / step_size)): pre_toks = word_map[i * step_size : i * step_size + passage_len] start_section_id = max([0] + [j for j in section_indices if j <= pre_toks[0][0]]) section_ids = [j for j in section_indices if j >= start_section_id and j <= pre_toks[-1][0]] section_ids = section_ids if len(section_ids) > 0 else [0] passage_text = " ".join([w for p_id, s_id, w in pre_toks]) passages += [ { "article_title": article_title, "section_title": " & ".join([paragraphs[j] for j in section_ids]), "wiki_id": article["wikidata_id"], "start_paragraph": pre_toks[0][0], "start_character": pre_toks[0][1], "end_paragraph": pre_toks[-1][0], "end_character": pre_toks[-1][1] + len(pre_toks[-1][2]) + 1, "passage_text": passage_text.replace("_NEWLINE_", "\n"), } ] return passages def wikipedia_article_snippets(article, passage_len=100, overlap=0): paragraphs = [par for par in article["text"].split("\n") if not par.startswith("Category:")] if "References" in paragraphs: paragraphs = paragraphs[: paragraphs.index("References")] article_title = article["title"] section_indices = [ i + 1 for i, par in enumerate(paragraphs[:-2]) if paragraphs[i] == "" and paragraphs[i + 1] != "" and paragraphs[i + 2] != "" ] par_tabs = [par.split(" ") for par in paragraphs] word_map = [(i, len(" ".join(par[:j])), w) for i, par in enumerate(par_tabs) for j, w in enumerate(par)] step_size = passage_len - overlap passages = [] for i in range(math.ceil(len(word_map) / step_size)): pre_toks = word_map[i * step_size : i * step_size + passage_len] start_section_id = max([0] + [j for j in section_indices if j <= pre_toks[0][0]]) section_ids = [j for j in section_indices if j >= start_section_id and j <= pre_toks[-1][0]] section_ids = section_ids if len(section_ids) > 0 else [-1] passage_text = " ".join([w for p_id, s_id, w in pre_toks]) passages += [ { "article_title": article_title, "section_title": " & ".join(["Start" if j == -1 else paragraphs[j].strip() for j in section_ids]), "wiki_id": article_title.replace(" ", "_"), "start_paragraph": pre_toks[0][0], "start_character": pre_toks[0][1], "end_paragraph": pre_toks[-1][0], "end_character": pre_toks[-1][1] + len(pre_toks[-1][2]) + 1, "passage_text": passage_text, } ] return passages _SPLIT_FUCNTION_MAP = { "wikipedia": wikipedia_article_snippets, "wiki40b": wiki40b_article_snippets, } def generate_snippets(wikipedia, split_funtion, passage_len=100, overlap=0): for i, article in enumerate(wikipedia): for doc in split_funtion(article, passage_len, overlap): part_id = json.dumps( { "datasets_id": i, "wiki_id": doc["wiki_id"], "sp": doc["start_paragraph"], "sc": doc["start_character"], "ep": doc["end_paragraph"], "ec": doc["end_character"], } ) doc["_id"] = part_id doc["datasets_id"] = i yield doc class WikiSnippetsConfig(datasets.BuilderConfig): """BuilderConfig for WikiSnippets.""" def __init__( self, wikipedia_name="wiki40b", wikipedia_version_name="en", snippets_length=100, snippets_overlap=0, **kwargs ): """BuilderConfig for WikiSnippets. Args: **kwargs: keyword arguments forwarded to super. """ super(WikiSnippetsConfig, self).__init__(**kwargs) self.wikipedia_name = wikipedia_name self.wikipedia_version_name = wikipedia_version_name self.snippets_length = snippets_length self.snippets_overlap = snippets_overlap class WikiSnippets(datasets.GeneratorBasedBuilder): BUILDER_CONFIG_CLASS = WikiSnippetsConfig BUILDER_CONFIGS = [ WikiSnippetsConfig( name="wiki40b_en_100_0", version=datasets.Version("1.0.0"), wikipedia_name="wiki40b", wikipedia_version_name="en", snippets_length=100, snippets_overlap=0, ), WikiSnippetsConfig( name="wikipedia_en_100_0", version=datasets.Version("1.0.0"), wikipedia_name="wikipedia", wikipedia_version_name="20200501.en", snippets_length=100, snippets_overlap=0, ), ] test_dummy_data = False def _info(self): return datasets.DatasetInfo( description=_DESCRIPTION, features=datasets.Features( { "_id": datasets.Value("string"), "datasets_id": datasets.Value("int32"), "wiki_id": datasets.Value("string"), "start_paragraph": datasets.Value("int32"), "start_character": datasets.Value("int32"), "end_paragraph": datasets.Value("int32"), "end_character": datasets.Value("int32"), "article_title": datasets.Value("string"), "section_title": datasets.Value("string"), "passage_text": datasets.Value("string"), } ), supervised_keys=None, homepage="https://dumps.wikimedia.org", citation=_CITATION, ) def _split_generators(self, dl_manager): wikipedia = datasets.load_dataset( path=self.config.wikipedia_name, name=self.config.wikipedia_version_name, ) return [ datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"wikipedia": wikipedia}), ] def _generate_examples(self, wikipedia): logger.info( "generating examples from = {} {}".format(self.config.wikipedia_name, self.config.wikipedia_version_name) ) for split in wikipedia: dset = wikipedia[split] split_function = _SPLIT_FUCNTION_MAP[self.config.wikipedia_name] for doc in generate_snippets( dset, split_function, passage_len=self.config.snippets_length, overlap=self.config.snippets_overlap ): id_ = doc["_id"] yield id_, doc