Update files from the datasets library (from 1.12.0)

Release notes: https://github.com/huggingface/datasets/releases/tag/1.12.0

dataset_infos.json

@@ -0,0 +1 @@
+{"all": {"description": "WikiHow is a new large-scale dataset using the online WikiHow\n(http://www.wikihow.com/) knowledge base.\n\nThere are two features:\n  - text: wikihow answers texts.\n  - headline: bold lines as summary.\n\nThere are two separate versions:\n  - all: consisting of the concatenation of all paragraphs as the articles and\n         the bold lines as the reference summaries.\n  - sep: consisting of each paragraph and its summary.\n\nDownload \"wikihowAll.csv\" and \"wikihowSep.csv\" from\nhttps://github.com/mahnazkoupaee/WikiHow-Dataset and place them in manual folder\nhttps://www.tensorflow.org/datasets/api_docs/python/tfds/download/DownloadConfig.\nTrain/validation/test splits are provided by the authors.\nPreprocessing is applied to remove short articles\n(abstract length < 0.75 article length) and clean up extra commas.\n", "citation": "@misc{koupaee2018wikihow,\n    title={WikiHow: A Large Scale Text Summarization Dataset},\n    author={Mahnaz Koupaee and William Yang Wang},\n    year={2018},\n    eprint={1810.09305},\n    archivePrefix={arXiv},\n    primaryClass={cs.CL}\n}\n", "homepage": "https://github.com/mahnazkoupaee/WikiHow-Dataset", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "headline": {"dtype": "string", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "wikihow", "config_name": "all", "version": {"version_str": "1.2.0", "description": null, "major": 1, "minor": 2, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 513238309, "num_examples": 157252, "dataset_name": "wikihow"}, "validation": {"name": "validation", "num_bytes": 18246897, "num_examples": 5599, "dataset_name": "wikihow"}, "test": {"name": "test", "num_bytes": 18276023, "num_examples": 5577, "dataset_name": "wikihow"}}, "download_checksums": {"https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_train.txt": {"num_bytes": 5096074, "checksum": "149bdd47ca9607bc5c730805d01c8fd879ffd7f328fd6869e6109288e8cfe733"}, "https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_val.txt": {"num_bytes": 182165, "checksum": "759789579a0a96783f054387d5c2d7b537db75f462629116743f0ac3e7450be4"}, "https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_test.txt": {"num_bytes": 182146, "checksum": "c7b3037410bdfbed258cf96dcd5e7d04bc7432b9c20c6e6cd8063b6db84f8f94"}}, "download_size": 5460385, "post_processing_size": null, "dataset_size": 549761229, "size_in_bytes": 555221614}, "sep": {"description": "WikiHow is a new large-scale dataset using the online WikiHow\n(http://www.wikihow.com/) knowledge base.\n\nThere are two features:\n  - text: wikihow answers texts.\n  - headline: bold lines as summary.\n\nThere are two separate versions:\n  - all: consisting of the concatenation of all paragraphs as the articles and\n         the bold lines as the reference summaries.\n  - sep: consisting of each paragraph and its summary.\n\nDownload \"wikihowAll.csv\" and \"wikihowSep.csv\" from\nhttps://github.com/mahnazkoupaee/WikiHow-Dataset and place them in manual folder\nhttps://www.tensorflow.org/datasets/api_docs/python/tfds/download/DownloadConfig.\nTrain/validation/test splits are provided by the authors.\nPreprocessing is applied to remove short articles\n(abstract length < 0.75 article length) and clean up extra commas.\n", "citation": "@misc{koupaee2018wikihow,\n    title={WikiHow: A Large Scale Text Summarization Dataset},\n    author={Mahnaz Koupaee and William Yang Wang},\n    year={2018},\n    eprint={1810.09305},\n    archivePrefix={arXiv},\n    primaryClass={cs.CL}\n}\n", "homepage": "https://github.com/mahnazkoupaee/WikiHow-Dataset", "license": "", "features": {"text": {"dtype": "string", "id": null, "_type": "Value"}, "headline": {"dtype": "string", "id": null, "_type": "Value"}, "title": {"dtype": "string", "id": null, "_type": "Value"}, "overview": {"dtype": "string", "id": null, "_type": "Value"}, "sectionLabel": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "wikihow", "config_name": "sep", "version": {"version_str": "1.2.0", "description": null, "major": 1, "minor": 2, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 990499776, "num_examples": 1060732, "dataset_name": "wikihow"}, "validation": {"name": "validation", "num_bytes": 35173966, "num_examples": 37932, "dataset_name": "wikihow"}, "test": {"name": "test", "num_bytes": 35271826, "num_examples": 37800, "dataset_name": "wikihow"}}, "download_checksums": {"https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_train.txt": {"num_bytes": 5096074, "checksum": "149bdd47ca9607bc5c730805d01c8fd879ffd7f328fd6869e6109288e8cfe733"}, "https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_val.txt": {"num_bytes": 182165, "checksum": "759789579a0a96783f054387d5c2d7b537db75f462629116743f0ac3e7450be4"}, "https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_test.txt": {"num_bytes": 182146, "checksum": "c7b3037410bdfbed258cf96dcd5e7d04bc7432b9c20c6e6cd8063b6db84f8f94"}}, "download_size": 5460385, "post_processing_size": null, "dataset_size": 1060945568, "size_in_bytes": 1066405953}}

urls_checksums/checksums.txt

@@ -1,3 +0,0 @@
-https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_test.txt 182146 c7b3037410bdfbed258cf96dcd5e7d04bc7432b9c20c6e6cd8063b6db84f8f94
-https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_train.txt 5096074 149bdd47ca9607bc5c730805d01c8fd879ffd7f328fd6869e6109288e8cfe733
-https://raw.githubusercontent.com/mahnazkoupaee/WikiHow-Dataset/master/all_val.txt 182165 759789579a0a96783f054387d5c2d7b537db75f462629116743f0ac3e7450be4

wikihow.py

@@ -24,7 +24,7 @@ import re
 import datasets
 
 
-_CITATION = """
+_CITATION = """\
 @misc{koupaee2018wikihow,
     title={WikiHow: A Large Scale Text Summarization Dataset},
     author={Mahnaz Koupaee and William Yang Wang},
@@ -35,7 +35,7 @@ _CITATION = """
 }
 """
 
-_DESCRIPTION = """
+_DESCRIPTION = """\
 WikiHow is a new large-scale dataset using the online WikiHow
 (http://www.wikihow.com/) knowledge base.