Oleg Somov
commited on
Commit
•
e560e57
1
Parent(s):
4392fcb
delete load script
Browse files
pauq.py
DELETED
@@ -1,285 +0,0 @@
|
|
1 |
-
# coding=utf-8
|
2 |
-
# 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 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 |
-
"""PAUQ: Text-to-SQL in Russian"""
|
16 |
-
|
17 |
-
import json
|
18 |
-
import os
|
19 |
-
|
20 |
-
import datasets
|
21 |
-
|
22 |
-
logger = datasets.logging.get_logger(__name__)
|
23 |
-
|
24 |
-
_CITATION = """\
|
25 |
-
@inproceedings{bakshandaeva-etal-2022-pauq,
|
26 |
-
title = "{PAUQ}: Text-to-{SQL} in {R}ussian",
|
27 |
-
author = "Bakshandaeva, Daria and
|
28 |
-
Somov, Oleg and
|
29 |
-
Dmitrieva, Ekaterina and
|
30 |
-
Davydova, Vera and
|
31 |
-
Tutubalina, Elena",
|
32 |
-
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
|
33 |
-
month = dec,
|
34 |
-
year = "2022",
|
35 |
-
address = "Abu Dhabi, United Arab Emirates",
|
36 |
-
publisher = "Association for Computational Linguistics",
|
37 |
-
url = "https://aclanthology.org/2022.findings-emnlp.175",
|
38 |
-
"""
|
39 |
-
|
40 |
-
_DESCRIPTION = """\
|
41 |
-
Pauq is a first Russian text-to-SQL dataset translated from original Spider dataset
|
42 |
-
with corrections and refinements of question, queries and databases.
|
43 |
-
"""
|
44 |
-
|
45 |
-
_LICENSE = "CC BY-SA 4.0"
|
46 |
-
|
47 |
-
_HOMEPAGE = "https://github.com/ai-spiderweb/pauq"
|
48 |
-
|
49 |
-
_URL = "https://huggingface.co/datasets/composite/pauq/resolve/main/formatted_pauq.zip"
|
50 |
-
|
51 |
-
RUSSIAN_PAUQ_TRL_DESCRIPTION = "Russian PAUQ train/test split based on target length of SQL query. Long query template in train, short query template in test."
|
52 |
-
ENGLISH_PAUQ_TRL_DESCRIPTION = "English PAUQ train/test split based on target length of SQL query. Long query template in train, short query template in test."
|
53 |
-
|
54 |
-
RUSSIAN_PAUQ_TSL_DESCRIPTION = "Russian PAUQ train/test split based on target length of SQL query. Short query template in train, long query template in test."
|
55 |
-
ENGLISH_PAUQ_TSL_DESCRIPTION = "English PAUQ train/test split based on target length of SQL query. Short query template in train, long query template in test."
|
56 |
-
|
57 |
-
RUSSIAN_PAUQ_OS_DESCRIPTION = "Independent and identical Russian PAUQ train/test split. Сorresponds to original Spider splitting."
|
58 |
-
ENGLISH_PAUQ_OS_DESCRIPTION = "Independent and identical English PAUQ train/test split. Сorresponds to original Spider splitting."
|
59 |
-
|
60 |
-
|
61 |
-
class Pauq(datasets.GeneratorBasedBuilder):
|
62 |
-
VERSION = datasets.Version("1.0.0")
|
63 |
-
|
64 |
-
BUILDER_CONFIGS = [
|
65 |
-
datasets.BuilderConfig(
|
66 |
-
name="ru_trl",
|
67 |
-
version=VERSION,
|
68 |
-
description=RUSSIAN_PAUQ_TRL_DESCRIPTION,
|
69 |
-
),
|
70 |
-
datasets.BuilderConfig(
|
71 |
-
name="en_trl",
|
72 |
-
version=VERSION,
|
73 |
-
description=ENGLISH_PAUQ_TRL_DESCRIPTION,
|
74 |
-
),
|
75 |
-
datasets.BuilderConfig(
|
76 |
-
name="ru_tsl",
|
77 |
-
version=VERSION,
|
78 |
-
description=RUSSIAN_PAUQ_TSL_DESCRIPTION,
|
79 |
-
),
|
80 |
-
datasets.BuilderConfig(
|
81 |
-
name="en_tsl",
|
82 |
-
version=VERSION,
|
83 |
-
description=ENGLISH_PAUQ_TSL_DESCRIPTION,
|
84 |
-
),
|
85 |
-
datasets.BuilderConfig(
|
86 |
-
name="ru_os",
|
87 |
-
version=VERSION,
|
88 |
-
description=RUSSIAN_PAUQ_OS_DESCRIPTION,
|
89 |
-
),
|
90 |
-
datasets.BuilderConfig(
|
91 |
-
name="en_os",
|
92 |
-
version=VERSION,
|
93 |
-
description=ENGLISH_PAUQ_OS_DESCRIPTION,
|
94 |
-
),
|
95 |
-
]
|
96 |
-
|
97 |
-
def _info(self):
|
98 |
-
features = datasets.Features(
|
99 |
-
{
|
100 |
-
"id": datasets.Value("string"),
|
101 |
-
"db_id": datasets.Value("string"),
|
102 |
-
"source": datasets.Value("string"),
|
103 |
-
"type": datasets.Value("string"),
|
104 |
-
"question": datasets.Value("string"),
|
105 |
-
"query": datasets.Value("string"),
|
106 |
-
"sql": datasets.features.Sequence(datasets.Value("string")),
|
107 |
-
"question_toks": datasets.features.Sequence(datasets.Value("string")),
|
108 |
-
"query_toks": datasets.features.Sequence(datasets.Value("string")),
|
109 |
-
"query_toks_no_values": datasets.features.Sequence(datasets.Value("string")),
|
110 |
-
"template": datasets.Value("string")
|
111 |
-
}
|
112 |
-
)
|
113 |
-
dataset_info = None
|
114 |
-
if self.config.name == 'ru_trl':
|
115 |
-
dataset_info = datasets.DatasetInfo(
|
116 |
-
description=RUSSIAN_PAUQ_TRL_DESCRIPTION,
|
117 |
-
features=features,
|
118 |
-
supervised_keys=None,
|
119 |
-
homepage=_HOMEPAGE,
|
120 |
-
license=_LICENSE,
|
121 |
-
citation=_CITATION,
|
122 |
-
config_name="ru_trl")
|
123 |
-
elif self.config.name == "en_trl":
|
124 |
-
dataset_info = datasets.DatasetInfo(
|
125 |
-
description=ENGLISH_PAUQ_TRL_DESCRIPTION,
|
126 |
-
features=features,
|
127 |
-
supervised_keys=None,
|
128 |
-
homepage=_HOMEPAGE,
|
129 |
-
license=_LICENSE,
|
130 |
-
citation=_CITATION,
|
131 |
-
config_name="en_trl")
|
132 |
-
elif self.config.name == 'ru_os':
|
133 |
-
dataset_info = datasets.DatasetInfo(
|
134 |
-
description=RUSSIAN_PAUQ_OS_DESCRIPTION,
|
135 |
-
features=features,
|
136 |
-
supervised_keys=None,
|
137 |
-
homepage=_HOMEPAGE,
|
138 |
-
license=_LICENSE,
|
139 |
-
citation=_CITATION,
|
140 |
-
config_name="ru_os")
|
141 |
-
elif self.config.name == 'en_os':
|
142 |
-
dataset_info = datasets.DatasetInfo(
|
143 |
-
description=ENGLISH_PAUQ_OS_DESCRIPTION,
|
144 |
-
features=features,
|
145 |
-
supervised_keys=None,
|
146 |
-
homepage=_HOMEPAGE,
|
147 |
-
license=_LICENSE,
|
148 |
-
citation=_CITATION,
|
149 |
-
config_name="en_os")
|
150 |
-
elif self.config.name == 'ru_tsl':
|
151 |
-
dataset_info = datasets.DatasetInfo(
|
152 |
-
description=RUSSIAN_PAUQ_TSL_DESCRIPTION,
|
153 |
-
features=features,
|
154 |
-
supervised_keys=None,
|
155 |
-
homepage=_HOMEPAGE,
|
156 |
-
license=_LICENSE,
|
157 |
-
citation=_CITATION,
|
158 |
-
config_name="ru_tsl")
|
159 |
-
elif self.config.name == "en_tsl":
|
160 |
-
dataset_info = datasets.DatasetInfo(
|
161 |
-
description=ENGLISH_PAUQ_TSL_DESCRIPTION,
|
162 |
-
features=features,
|
163 |
-
supervised_keys=None,
|
164 |
-
homepage=_HOMEPAGE,
|
165 |
-
license=_LICENSE,
|
166 |
-
citation=_CITATION,
|
167 |
-
config_name="en_tsl")
|
168 |
-
|
169 |
-
return dataset_info
|
170 |
-
|
171 |
-
def _split_generators(self, dl_manager):
|
172 |
-
downloaded_filepath = dl_manager.download_and_extract(_URL)
|
173 |
-
|
174 |
-
dataset_name = self.config.name
|
175 |
-
|
176 |
-
splits = []
|
177 |
-
if dataset_name == 'ru_trl':
|
178 |
-
splits = [
|
179 |
-
datasets.SplitGenerator(
|
180 |
-
name=datasets.Split.TRAIN,
|
181 |
-
gen_kwargs={
|
182 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_trl_train.json"),
|
183 |
-
},
|
184 |
-
),
|
185 |
-
datasets.SplitGenerator(
|
186 |
-
name=datasets.Split.TEST,
|
187 |
-
gen_kwargs={
|
188 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_trl_test.json"),
|
189 |
-
},
|
190 |
-
)
|
191 |
-
]
|
192 |
-
elif dataset_name == 'en_trl':
|
193 |
-
splits = [
|
194 |
-
datasets.SplitGenerator(
|
195 |
-
name=datasets.Split.TRAIN,
|
196 |
-
gen_kwargs={
|
197 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_trl_train.json"),
|
198 |
-
},
|
199 |
-
),
|
200 |
-
datasets.SplitGenerator(
|
201 |
-
name=datasets.Split.TEST,
|
202 |
-
gen_kwargs={
|
203 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_trl_test.json"),
|
204 |
-
},
|
205 |
-
)]
|
206 |
-
elif dataset_name == 'ru_os':
|
207 |
-
splits = [
|
208 |
-
datasets.SplitGenerator(
|
209 |
-
name=datasets.Split.TRAIN,
|
210 |
-
gen_kwargs={
|
211 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_os_train.json"),
|
212 |
-
},
|
213 |
-
),
|
214 |
-
datasets.SplitGenerator(
|
215 |
-
name=datasets.Split.TEST,
|
216 |
-
gen_kwargs={
|
217 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_os_test.json"),
|
218 |
-
},
|
219 |
-
)
|
220 |
-
]
|
221 |
-
elif dataset_name == 'en_os':
|
222 |
-
splits = [
|
223 |
-
datasets.SplitGenerator(
|
224 |
-
name=datasets.Split.TRAIN,
|
225 |
-
gen_kwargs={
|
226 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_os_train.json"),
|
227 |
-
},
|
228 |
-
),
|
229 |
-
datasets.SplitGenerator(
|
230 |
-
name=datasets.Split.TEST,
|
231 |
-
gen_kwargs={
|
232 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_os_test.json"),
|
233 |
-
},
|
234 |
-
)
|
235 |
-
]
|
236 |
-
elif dataset_name == 'ru_tsl':
|
237 |
-
splits = [
|
238 |
-
datasets.SplitGenerator(
|
239 |
-
name=datasets.Split.TRAIN,
|
240 |
-
gen_kwargs={
|
241 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_tsl_train.json"),
|
242 |
-
},
|
243 |
-
),
|
244 |
-
datasets.SplitGenerator(
|
245 |
-
name=datasets.Split.TEST,
|
246 |
-
gen_kwargs={
|
247 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/ru_tsl_test.json"),
|
248 |
-
},
|
249 |
-
)
|
250 |
-
]
|
251 |
-
elif dataset_name == 'en_tsl':
|
252 |
-
splits = [
|
253 |
-
datasets.SplitGenerator(
|
254 |
-
name=datasets.Split.TRAIN,
|
255 |
-
gen_kwargs={
|
256 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_tsl_train.json"),
|
257 |
-
},
|
258 |
-
),
|
259 |
-
datasets.SplitGenerator(
|
260 |
-
name=datasets.Split.TEST,
|
261 |
-
gen_kwargs={
|
262 |
-
"data_filepath": os.path.join(downloaded_filepath, "formatted_pauq/splits/en_tsl_test.json"),
|
263 |
-
},
|
264 |
-
)]
|
265 |
-
return splits
|
266 |
-
|
267 |
-
def _generate_examples(self, data_filepath):
|
268 |
-
"""This function returns the examples in the raw (text) form."""
|
269 |
-
logger.info("generating examples from = %s", data_filepath)
|
270 |
-
with open(data_filepath, encoding="utf-8") as f:
|
271 |
-
pauq = json.load(f)
|
272 |
-
for idx, sample in enumerate(pauq):
|
273 |
-
yield idx, {
|
274 |
-
"id": sample["id"],
|
275 |
-
"db_id": sample["db_id"],
|
276 |
-
"source": sample["source"],
|
277 |
-
"type": sample["type"],
|
278 |
-
"query": sample["query"],
|
279 |
-
"sql": sample['sql'],
|
280 |
-
"question": sample["question"],
|
281 |
-
"question_toks": sample["question_toks"],
|
282 |
-
"query_toks": sample["query_toks"],
|
283 |
-
"query_toks_no_values": sample["query_toks_no_values"],
|
284 |
-
"template": sample["template"]
|
285 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|