Creating custom loading script.
Browse files
README.md
CHANGED
@@ -6,19 +6,6 @@ language:
|
|
6 |
- pt
|
7 |
size_categories:
|
8 |
- 1M<n<10M
|
9 |
-
configs:
|
10 |
-
- config_name: quati_10M_passages
|
11 |
-
data_files: quati_10M_part_*.tsv
|
12 |
-
sep: "\t"
|
13 |
-
- config_name: quati_1M_passages
|
14 |
-
data_files: quati_1M.tsv
|
15 |
-
sep: "\t"
|
16 |
-
- config_name: topics
|
17 |
-
data_files: topics/*.tsv
|
18 |
-
sep: "\t"
|
19 |
-
- config_name: qrels
|
20 |
-
data_files: qrels/*
|
21 |
-
sep: " "
|
22 |
---
|
23 |
|
24 |
# Quati Information Retrieval Dataset
|
|
|
6 |
- pt
|
7 |
size_categories:
|
8 |
- 1M<n<10M
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
# Quati Information Retrieval Dataset
|
quati.py
ADDED
@@ -0,0 +1,202 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
"""Quati dataset."""
|
18 |
+
|
19 |
+
import datasets
|
20 |
+
|
21 |
+
|
22 |
+
_CITATION = """
|
23 |
+
place holder
|
24 |
+
"""
|
25 |
+
|
26 |
+
_URL = "https://github.com/unicamp-dl/quati"
|
27 |
+
|
28 |
+
_DESCRIPTION = """
|
29 |
+
Quati ― Portuguese Native Information Retrieval dataset.
|
30 |
+
"""
|
31 |
+
|
32 |
+
|
33 |
+
|
34 |
+
QUATI_10M_DATASET_PARTS=["part_00", "part_01", "part_02", "part_03", "part_04"]
|
35 |
+
|
36 |
+
|
37 |
+
|
38 |
+
_URLS = {
|
39 |
+
"quati_1M_passages": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_1M_passages.tsv",
|
40 |
+
"quati_10M_passages_part_00": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_10M_part_00.tsv",
|
41 |
+
"quati_10M_passages_part_01": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_10M_part_01.tsv",
|
42 |
+
"quati_10M_passages_part_02": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_10M_part_02.tsv",
|
43 |
+
"quati_10M_passages_part_03": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_10M_part_03.tsv",
|
44 |
+
"quati_10M_passages_part_04": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/quati_10M_part_04.tsv",
|
45 |
+
"quati_1M_qrels": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/qrels/quati_1M_qrels.txt",
|
46 |
+
"quati_10M_qrels": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/qrels/quati_10M_qrels.txt",
|
47 |
+
"quati_test_topics": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/topics/quati_test_topics.txt",
|
48 |
+
"quati_all_topics": "https://huggingface.co/datasets/unicamp-dl/quati/resolve/main/topics/quati_all_topics.txt"
|
49 |
+
}
|
50 |
+
|
51 |
+
|
52 |
+
|
53 |
+
def generate_examples_passages(filepath):
|
54 |
+
|
55 |
+
with open(filepath, encoding="utf-8") as input_file:
|
56 |
+
for (idx, line) in enumerate(input_file):
|
57 |
+
|
58 |
+
passage_id, passage = line.rstrip().split("\t")
|
59 |
+
|
60 |
+
features = {"passage_id": passage_id,
|
61 |
+
"passage": passage}
|
62 |
+
|
63 |
+
yield idx, features
|
64 |
+
|
65 |
+
|
66 |
+
|
67 |
+
def generate_examples_qrels(filepath):
|
68 |
+
|
69 |
+
with open(filepath, encoding="utf-8") as input_file:
|
70 |
+
for (idx, line) in enumerate(input_file):
|
71 |
+
query_id, _, passage_id, score = line.rstrip().split(" ")
|
72 |
+
|
73 |
+
features = {"query_id": int(query_id),
|
74 |
+
"passage_id": passage_id,
|
75 |
+
"score": int(score)}
|
76 |
+
|
77 |
+
yield idx, features
|
78 |
+
|
79 |
+
|
80 |
+
def generate_examples_topics(filepath):
|
81 |
+
|
82 |
+
with open(filepath, encoding="utf-8") as input_file:
|
83 |
+
for (idx, line) in enumerate(input_file):
|
84 |
+
query_id, query = line.rstrip().split("\t")
|
85 |
+
|
86 |
+
features = {"query_id": int(query_id),
|
87 |
+
"query": query}
|
88 |
+
|
89 |
+
yield idx, features
|
90 |
+
|
91 |
+
|
92 |
+
|
93 |
+
class Quati(datasets.GeneratorBasedBuilder):
|
94 |
+
|
95 |
+
BUILDER_CONFIGS = (
|
96 |
+
[
|
97 |
+
datasets.BuilderConfig(
|
98 |
+
name="quati_10M_passages",
|
99 |
+
description="Portugues Brazilian passages, composing the complete Quati 10M dataset.",
|
100 |
+
version=datasets.Version("1.0.0"),
|
101 |
+
),
|
102 |
+
|
103 |
+
datasets.BuilderConfig(
|
104 |
+
name="quati_1M_passages",
|
105 |
+
description="Portugues Brazilian passages, composing the Quati 1M dataset.",
|
106 |
+
version=datasets.Version("1.0.0"),
|
107 |
+
),
|
108 |
+
|
109 |
+
datasets.BuilderConfig(
|
110 |
+
name="quati_10M_qrels",
|
111 |
+
description="Qrels for the annotated passages from the Quati 10M dataset.",
|
112 |
+
version=datasets.Version("1.0.0"),
|
113 |
+
),
|
114 |
+
|
115 |
+
datasets.BuilderConfig(
|
116 |
+
name="quati_1M_qrels",
|
117 |
+
description="Qrels for the annotated passages from the Quati 1M dataset.",
|
118 |
+
version=datasets.Version("1.0.0"),
|
119 |
+
),
|
120 |
+
|
121 |
+
datasets.BuilderConfig(
|
122 |
+
name="quati_test_topics",
|
123 |
+
description="50 test topics, corresponding to Quati dataset qrels.",
|
124 |
+
version=datasets.Version("1.0.0"),
|
125 |
+
),
|
126 |
+
|
127 |
+
datasets.BuilderConfig(
|
128 |
+
name="quati_all_topics",
|
129 |
+
description="All 200 topics created for the Quati dataset, including the 50 ones corresponding to Quati dataset qrels.",
|
130 |
+
version=datasets.Version("1.0.0"),
|
131 |
+
)
|
132 |
+
]
|
133 |
+
+ [
|
134 |
+
datasets.BuilderConfig(
|
135 |
+
name="quati_10M_passages_{}".format(which_part),
|
136 |
+
description="Portugues Brazilian passages, composing the Quati 10M dataset {}.".format(which_part),
|
137 |
+
version=datasets.Version("1.0.0"),
|
138 |
+
)
|
139 |
+
for which_part in QUATI_10M_DATASET_PARTS
|
140 |
+
]
|
141 |
+
)
|
142 |
+
|
143 |
+
DEFAULT_CONFIG_NAME = "quati_1M_passages"
|
144 |
+
|
145 |
+
|
146 |
+
def _info(self):
|
147 |
+
name = self.config.name
|
148 |
+
if "passages" in name:
|
149 |
+
features = {
|
150 |
+
"passage_id": datasets.Value("string"),
|
151 |
+
"passage": datasets.Value("string"),
|
152 |
+
}
|
153 |
+
elif name.endswith("qrels"):
|
154 |
+
features = {
|
155 |
+
"query_id": datasets.Value("int32"),
|
156 |
+
"passage_id": datasets.Value("string"),
|
157 |
+
"score": datasets.Value("int32"),
|
158 |
+
}
|
159 |
+
else:
|
160 |
+
features = {
|
161 |
+
"query_id": datasets.Value("int32"),
|
162 |
+
"query": datasets.Value("string"),
|
163 |
+
}
|
164 |
+
|
165 |
+
return datasets.DatasetInfo(
|
166 |
+
description=f"{_DESCRIPTION}\n{self.config.description}",
|
167 |
+
features=datasets.Features(features),
|
168 |
+
supervised_keys=None,
|
169 |
+
homepage=_URL,
|
170 |
+
citation=_CITATION,
|
171 |
+
)
|
172 |
+
|
173 |
+
|
174 |
+
def _split_generators(self, dl_manager):
|
175 |
+
"""Returns SplitGenerators."""
|
176 |
+
|
177 |
+
if self.config.name == "quati_10M_passages":
|
178 |
+
|
179 |
+
urls = {which_part: _URLS["quati_10M_passages_{}".format(which_part)] for which_part in QUATI_10M_DATASET_PARTS}
|
180 |
+
|
181 |
+
dl_path = dl_manager.download_and_extract(urls)
|
182 |
+
|
183 |
+
return [datasets.SplitGenerator(name="quati_10M_passages_{}".format(which_part), gen_kwargs={"filepath": dl_path[which_part]}) for which_part in QUATI_10M_DATASET_PARTS]
|
184 |
+
|
185 |
+
else:
|
186 |
+
url = _URLS[self.config.name]
|
187 |
+
dl_path = dl_manager.download_and_extract(url)
|
188 |
+
|
189 |
+
return (datasets.SplitGenerator(name=self.config.name, gen_kwargs={"filepath": dl_path}),)
|
190 |
+
|
191 |
+
|
192 |
+
def _generate_examples(self, filepath, args=None):
|
193 |
+
"""Yields examples."""
|
194 |
+
|
195 |
+
if "passages" in self.config.name:
|
196 |
+
return generate_examples_passages(filepath)
|
197 |
+
|
198 |
+
if self.config.name.endswith("qrels"):
|
199 |
+
return generate_examples_qrels(filepath)
|
200 |
+
|
201 |
+
else:
|
202 |
+
return generate_examples_topics(filepath)
|