Datasets:
Tasks:
Question Answering
Sub-tasks:
extractive-qa
Languages:
English
Size:
100K<n<1M
ArXiv:
License:
Commit
•
1a3d5e6
0
Parent(s):
Update files from the datasets library (from 1.0.0)
Browse filesRelease notes: https://github.com/huggingface/datasets/releases/tag/1.0.0
- .gitattributes +27 -0
- dataset_infos.json +1 -0
- dummy/raw_jeopardy/1.0.0/dummy_data.zip +3 -0
- dummy/train_test_val/1.0.0/dummy_data.zip +3 -0
- search_qa.py +197 -0
.gitattributes
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
4 |
+
*.bin.* filter=lfs diff=lfs merge=lfs -text
|
5 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
11 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
12 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
13 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
14 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
15 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
16 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
17 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
18 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
19 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
20 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
21 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
22 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
23 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
24 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
26 |
+
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
dataset_infos.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"raw_jeopardy": {"description": "\n# pylint: disable=line-too-long\nWe publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind \nCNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect a full pipeline of general question-answering. That is, we start not from an existing article \nand generate a question-answer pair, but start from an existing question-answer pair, crawled from J! Archive, and augment it with text snippets retrieved by Google. \nFollowing this approach, we built SearchQA, which consists of more than 140k question-answer pairs with each pair having 49.6 snippets on average. Each question-answer-context\n tuple of the SearchQA comes with additional meta-data such as the snippet's URL, which we believe will be valuable resources for future research. We conduct human evaluation \n as well as test two baseline methods, one simple word selection and the other deep learning based, on the SearchQA. We show that there is a meaningful gap between the human \n and machine performances. This suggests that the proposed dataset could well serve as a benchmark for question-answering.\n\n", "citation": "\n @article{DBLP:journals/corr/DunnSHGCC17,\n author = {Matthew Dunn and\n Levent Sagun and\n Mike Higgins and\n V. Ugur G{\"{u}}ney and\n Volkan Cirik and\n Kyunghyun Cho},\n title = {SearchQA: {A} New Q{\\&}A Dataset Augmented with Context from a\n Search Engine},\n journal = {CoRR},\n volume = {abs/1704.05179},\n year = {2017},\n url = {http://arxiv.org/abs/1704.05179},\n archivePrefix = {arXiv},\n eprint = {1704.05179},\n timestamp = {Mon, 13 Aug 2018 16:47:09 +0200},\n biburl = {https://dblp.org/rec/journals/corr/DunnSHGCC17.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n }\n\n", "homepage": "https://github.com/nyu-dl/dl4ir-searchQA", "license": "", "features": {"category": {"dtype": "string", "id": null, "_type": "Value"}, "air_date": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "value": {"dtype": "string", "id": null, "_type": "Value"}, "answer": {"dtype": "string", "id": null, "_type": "Value"}, "round": {"dtype": "string", "id": null, "_type": "Value"}, "show_number": {"dtype": "int32", "id": null, "_type": "Value"}, "search_results": {"feature": {"urls": {"dtype": "string", "id": null, "_type": "Value"}, "snippets": {"dtype": "string", "id": null, "_type": "Value"}, "titles": {"dtype": "string", "id": null, "_type": "Value"}, "related_links": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "supervised_keys": null, "builder_name": "search_qa", "config_name": "raw_jeopardy", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 7770972348, "num_examples": 216757, "dataset_name": "search_qa"}}, "download_checksums": {"https://drive.google.com/uc?export=download&id=1U7WdBpd9kJ85S7BbBhWUSiy9NnXrKdO6": {"num_bytes": 3314386157, "checksum": "daaf1ddbb0c34c49832f6c8c26c9d59222085d45c7740425ccad9e38a9232cb4"}}, "download_size": 3314386157, "dataset_size": 7770972348, "size_in_bytes": 11085358505}, "train_test_val": {"description": "\n# pylint: disable=line-too-long\nWe publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind \nCNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect a full pipeline of general question-answering. That is, we start not from an existing article \nand generate a question-answer pair, but start from an existing question-answer pair, crawled from J! Archive, and augment it with text snippets retrieved by Google. \nFollowing this approach, we built SearchQA, which consists of more than 140k question-answer pairs with each pair having 49.6 snippets on average. Each question-answer-context\n tuple of the SearchQA comes with additional meta-data such as the snippet's URL, which we believe will be valuable resources for future research. We conduct human evaluation \n as well as test two baseline methods, one simple word selection and the other deep learning based, on the SearchQA. We show that there is a meaningful gap between the human \n and machine performances. This suggests that the proposed dataset could well serve as a benchmark for question-answering.\n\n", "citation": "\n @article{DBLP:journals/corr/DunnSHGCC17,\n author = {Matthew Dunn and\n Levent Sagun and\n Mike Higgins and\n V. Ugur G{\"{u}}ney and\n Volkan Cirik and\n Kyunghyun Cho},\n title = {SearchQA: {A} New Q{\\&}A Dataset Augmented with Context from a\n Search Engine},\n journal = {CoRR},\n volume = {abs/1704.05179},\n year = {2017},\n url = {http://arxiv.org/abs/1704.05179},\n archivePrefix = {arXiv},\n eprint = {1704.05179},\n timestamp = {Mon, 13 Aug 2018 16:47:09 +0200},\n biburl = {https://dblp.org/rec/journals/corr/DunnSHGCC17.bib},\n bibsource = {dblp computer science bibliography, https://dblp.org}\n }\n\n", "homepage": "https://github.com/nyu-dl/dl4ir-searchQA", "license": "", "features": {"category": {"dtype": "string", "id": null, "_type": "Value"}, "air_date": {"dtype": "string", "id": null, "_type": "Value"}, "question": {"dtype": "string", "id": null, "_type": "Value"}, "value": {"dtype": "string", "id": null, "_type": "Value"}, "answer": {"dtype": "string", "id": null, "_type": "Value"}, "round": {"dtype": "string", "id": null, "_type": "Value"}, "show_number": {"dtype": "int32", "id": null, "_type": "Value"}, "search_results": {"feature": {"urls": {"dtype": "string", "id": null, "_type": "Value"}, "snippets": {"dtype": "string", "id": null, "_type": "Value"}, "titles": {"dtype": "string", "id": null, "_type": "Value"}, "related_links": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}}, "supervised_keys": null, "builder_name": "search_qa", "config_name": "train_test_val", "version": {"version_str": "1.0.0", "description": "", "datasets_version_to_prepare": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 5303005740, "num_examples": 151295, "dataset_name": "search_qa"}, "test": {"name": "test", "num_bytes": 1466749978, "num_examples": 43228, "dataset_name": "search_qa"}, "validation": {"name": "validation", "num_bytes": 740962715, "num_examples": 21613, "dataset_name": "search_qa"}}, "download_checksums": {"https://drive.google.com/uc?export=download&id=1aHPVfC5TrlnUjehtagVZoDfq4VccgaNT": {"num_bytes": 3148550732, "checksum": "1f547df8b00e919ba692ca8c133462d358a89ee6b15a8c65c40efe006ed6c4eb"}}, "download_size": 3148550732, "dataset_size": 7510718433, "size_in_bytes": 10659269165}}
|
dummy/raw_jeopardy/1.0.0/dummy_data.zip
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d64bfb623c7393d509f8e83ebcbc12bc28b1d7684cc22c74ee6dcc47bdd838a3
|
3 |
+
size 21595
|
dummy/train_test_val/1.0.0/dummy_data.zip
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:00f07138b140d8bcbced894ac18789ca4602672b8d21644b334cef21b7cb9958
|
3 |
+
size 8758
|
search_qa.py
ADDED
@@ -0,0 +1,197 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# coding=utf-8
|
2 |
+
# Copyright 2020 The TensorFlow Datasets Authors and the HuggingFace Datasets Authors.
|
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 at
|
7 |
+
#
|
8 |
+
# http://www.apache.org/licenses/LICENSE-2.0
|
9 |
+
#
|
10 |
+
# Unless required by applicable law or agreed to in writing, software
|
11 |
+
# 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 and
|
14 |
+
# limitations under the License.
|
15 |
+
|
16 |
+
# Lint as: python3
|
17 |
+
"""SEMPRE dataset."""
|
18 |
+
|
19 |
+
from __future__ import absolute_import, division, print_function
|
20 |
+
|
21 |
+
import json
|
22 |
+
import os
|
23 |
+
|
24 |
+
import datasets
|
25 |
+
|
26 |
+
|
27 |
+
_CITATION = r"""
|
28 |
+
@article{DBLP:journals/corr/DunnSHGCC17,
|
29 |
+
author = {Matthew Dunn and
|
30 |
+
Levent Sagun and
|
31 |
+
Mike Higgins and
|
32 |
+
V. Ugur G{\"{u}}ney and
|
33 |
+
Volkan Cirik and
|
34 |
+
Kyunghyun Cho},
|
35 |
+
title = {SearchQA: {A} New Q{\&}A Dataset Augmented with Context from a
|
36 |
+
Search Engine},
|
37 |
+
journal = {CoRR},
|
38 |
+
volume = {abs/1704.05179},
|
39 |
+
year = {2017},
|
40 |
+
url = {http://arxiv.org/abs/1704.05179},
|
41 |
+
archivePrefix = {arXiv},
|
42 |
+
eprint = {1704.05179},
|
43 |
+
timestamp = {Mon, 13 Aug 2018 16:47:09 +0200},
|
44 |
+
biburl = {https://dblp.org/rec/journals/corr/DunnSHGCC17.bib},
|
45 |
+
bibsource = {dblp computer science bibliography, https://dblp.org}
|
46 |
+
}
|
47 |
+
|
48 |
+
"""
|
49 |
+
# pylint: disable=line-too-long
|
50 |
+
_DESCRIPTION = """
|
51 |
+
We publicly release a new large-scale dataset, called SearchQA, for machine comprehension, or question-answering. Unlike recently released datasets, such as DeepMind
|
52 |
+
CNN/DailyMail and SQuAD, the proposed SearchQA was constructed to reflect a full pipeline of general question-answering. That is, we start not from an existing article
|
53 |
+
and generate a question-answer pair, but start from an existing question-answer pair, crawled from J! Archive, and augment it with text snippets retrieved by Google.
|
54 |
+
Following this approach, we built SearchQA, which consists of more than 140k question-answer pairs with each pair having 49.6 snippets on average. Each question-answer-context
|
55 |
+
tuple of the SearchQA comes with additional meta-data such as the snippet's URL, which we believe will be valuable resources for future research. We conduct human evaluation
|
56 |
+
as well as test two baseline methods, one simple word selection and the other deep learning based, on the SearchQA. We show that there is a meaningful gap between the human
|
57 |
+
and machine performances. This suggests that the proposed dataset could well serve as a benchmark for question-answering.
|
58 |
+
"""
|
59 |
+
|
60 |
+
_DL_URLS = {
|
61 |
+
"raw_jeopardy": "https://drive.google.com/uc?export=download&id=1U7WdBpd9kJ85S7BbBhWUSiy9NnXrKdO6",
|
62 |
+
"train_test_val": "https://drive.google.com/uc?export=download&id=1aHPVfC5TrlnUjehtagVZoDfq4VccgaNT",
|
63 |
+
}
|
64 |
+
# pylint: enable=line-too-long
|
65 |
+
|
66 |
+
|
67 |
+
class SearchQaConfig(datasets.BuilderConfig):
|
68 |
+
"""BuilderConfig for SearchQA."""
|
69 |
+
|
70 |
+
def __init__(self, data_url, **kwargs):
|
71 |
+
"""BuilderConfig for SearchQA
|
72 |
+
|
73 |
+
Args:
|
74 |
+
**kwargs: keyword arguments forwarded to super.
|
75 |
+
"""
|
76 |
+
super(SearchQaConfig, self).__init__(version=datasets.Version("1.0.0", ""), **kwargs)
|
77 |
+
self.data_url = data_url
|
78 |
+
|
79 |
+
|
80 |
+
class SearchQa(datasets.GeneratorBasedBuilder):
|
81 |
+
"""Search QA Dataset."""
|
82 |
+
|
83 |
+
BUILDER_CONFIGS = [SearchQaConfig(name=name, description="", data_url=_DL_URLS[name]) for name in _DL_URLS.keys()]
|
84 |
+
|
85 |
+
def _info(self):
|
86 |
+
return datasets.DatasetInfo(
|
87 |
+
description=_DESCRIPTION + "\n" + self.config.description,
|
88 |
+
features=datasets.Features(
|
89 |
+
{
|
90 |
+
"category": datasets.Value("string"),
|
91 |
+
"air_date": datasets.Value("string"),
|
92 |
+
"question": datasets.Value("string"),
|
93 |
+
"value": datasets.Value("string"),
|
94 |
+
"answer": datasets.Value("string"),
|
95 |
+
"round": datasets.Value("string"),
|
96 |
+
"show_number": datasets.Value("int32"),
|
97 |
+
"search_results": datasets.features.Sequence(
|
98 |
+
{
|
99 |
+
"urls": datasets.Value("string"),
|
100 |
+
"snippets": datasets.Value("string"),
|
101 |
+
"titles": datasets.Value("string"),
|
102 |
+
"related_links": datasets.Value("string"),
|
103 |
+
}
|
104 |
+
)
|
105 |
+
# These are the features of your dataset like images, labels ...
|
106 |
+
}
|
107 |
+
),
|
108 |
+
homepage="https://github.com/nyu-dl/dl4ir-searchQA",
|
109 |
+
citation=_CITATION,
|
110 |
+
)
|
111 |
+
|
112 |
+
def _split_generators(self, dl_manager):
|
113 |
+
"""Returns SplitGenerators."""
|
114 |
+
# TODO(jeopardy): Downloads the data and defines the splits
|
115 |
+
# dl_manager is a datasets.download.DownloadManager that can be used to
|
116 |
+
# download and extract URLs
|
117 |
+
|
118 |
+
if self.config.name == "raw_jeopardy":
|
119 |
+
filepath = dl_manager.download_and_extract(_DL_URLS["raw_jeopardy"])
|
120 |
+
sub_folders = sorted(os.listdir(os.path.join(filepath, "jeopardy")))
|
121 |
+
all_files = []
|
122 |
+
for zip_folder in sub_folders:
|
123 |
+
if "lock" in zip_folder:
|
124 |
+
continue
|
125 |
+
zip_folder_path = os.path.join(filepath, "jeopardy", zip_folder)
|
126 |
+
file_path = dl_manager.extract(zip_folder_path)
|
127 |
+
zip_folder = zip_folder.split(".")[0]
|
128 |
+
if os.path.isdir(os.path.join(file_path, zip_folder)):
|
129 |
+
file_path = os.path.join(file_path, zip_folder)
|
130 |
+
|
131 |
+
else:
|
132 |
+
# in some cases the subfolder name contains sapces as 050000 - 059999 and 050000-059999
|
133 |
+
parts = zip_folder.split("-")
|
134 |
+
zip_folder = parts[0] + " - " + parts[1]
|
135 |
+
if os.path.isdir(os.path.join(file_path, zip_folder)):
|
136 |
+
file_path = os.path.join(file_path, zip_folder)
|
137 |
+
|
138 |
+
files = sorted(os.listdir(file_path))
|
139 |
+
|
140 |
+
files_paths = [os.path.join(file_path, file) for file in files if "__MACOSX" not in file]
|
141 |
+
all_files.extend(files_paths)
|
142 |
+
|
143 |
+
return [
|
144 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepaths": all_files}),
|
145 |
+
]
|
146 |
+
elif self.config.name == "train_test_val":
|
147 |
+
filepath = dl_manager.download_and_extract(_DL_URLS["train_test_val"])
|
148 |
+
train_path = dl_manager.extract(os.path.join(filepath, "data_json", "train.zip"))
|
149 |
+
test_path = dl_manager.extract(os.path.join(filepath, "data_json", "test.zip"))
|
150 |
+
val_path = dl_manager.extract(os.path.join(filepath, "data_json", "val.zip"))
|
151 |
+
|
152 |
+
train_files = [os.path.join(train_path, file) for file in sorted(os.listdir(train_path))]
|
153 |
+
test_files = [os.path.join(test_path, file) for file in sorted(os.listdir(test_path))]
|
154 |
+
val_files = [os.path.join(val_path, file) for file in sorted(os.listdir(val_path))]
|
155 |
+
return [
|
156 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepaths": train_files}),
|
157 |
+
datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepaths": test_files}),
|
158 |
+
datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepaths": val_files}),
|
159 |
+
]
|
160 |
+
|
161 |
+
def _generate_examples(self, filepaths):
|
162 |
+
"""Yields examples."""
|
163 |
+
# TODO(searchQa): Yields (key, example) tuples from the dataset
|
164 |
+
for i, filepath in enumerate(filepaths):
|
165 |
+
with open(filepath, encoding="utf-8") as f:
|
166 |
+
|
167 |
+
data = json.load(f)
|
168 |
+
category = data["category"]
|
169 |
+
air_date = data["air_date"]
|
170 |
+
question = data["question"]
|
171 |
+
value = data["value"]
|
172 |
+
answer = data["answer"]
|
173 |
+
round_ = data["round"]
|
174 |
+
show_number = int(data["show_number"])
|
175 |
+
search_results = data["search_results"]
|
176 |
+
urls = [result["url"] for result in search_results]
|
177 |
+
snippets = [result["snippet"] for result in search_results]
|
178 |
+
titles = [result["title"] for result in search_results]
|
179 |
+
related_links = [
|
180 |
+
result["related_links"] if result["related_links"] else "" for result in search_results
|
181 |
+
]
|
182 |
+
yield i, {
|
183 |
+
"category": category,
|
184 |
+
"air_date": air_date,
|
185 |
+
"question": question,
|
186 |
+
"value": value,
|
187 |
+
"answer": answer,
|
188 |
+
"round": round_,
|
189 |
+
"category": category,
|
190 |
+
"show_number": show_number,
|
191 |
+
"search_results": {
|
192 |
+
"urls": urls,
|
193 |
+
"snippets": snippets,
|
194 |
+
"titles": titles,
|
195 |
+
"related_links": related_links,
|
196 |
+
},
|
197 |
+
}
|