Datasets:

Languages:
Indonesian
ArXiv:
License:
holylovenia commited on
Commit
1b825db
1 Parent(s): d1098f5

Upload indowiki.py with huggingface_hub

Browse files
Files changed (1) hide show
  1. indowiki.py +198 -0
indowiki.py ADDED
@@ -0,0 +1,198 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # coding=utf-8
2
+ # Copyright 2022 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
+
16
+
17
+ from pathlib import Path
18
+ from typing import Dict, List, Tuple
19
+
20
+ import datasets
21
+
22
+ from seacrowd.utils.configs import SEACrowdConfig
23
+ from seacrowd.utils.constants import Licenses
24
+
25
+ _CITATION = """\
26
+ @INPROCEEDINGS{ramli2022indokepler,
27
+ author={Ramli, Inigo and Krisnadhi, Adila Alfa and Prasojo, Radityo Eko},
28
+ booktitle={2022 7th International Workshop on Big Data and Information Security (IWBIS)},
29
+ title={IndoKEPLER, IndoWiki, and IndoLAMA: A Knowledge-enhanced Language Model, Dataset, and Benchmark for the Indonesian Language},
30
+ year={2022},
31
+ volume={},
32
+ number={},
33
+ pages={19-26},
34
+ doi={10.1109/IWBIS56557.2022.9924844}}
35
+ """
36
+
37
+ _DATASETNAME = "indowiki"
38
+ _DESCRIPTION = """\
39
+ IndoWiki is a knowledge-graph dataset taken from WikiData and aligned with Wikipedia Bahasa Indonesia as it's corpus.
40
+ """
41
+ _HOMEPAGE = "https://github.com/IgoRamli/IndoWiki"
42
+ _LANGUAGES = ["ind"]
43
+ _LICENSE = Licenses.MIT.value
44
+ _LOCAL = False
45
+
46
+ _URLS = {
47
+ "inductive": {
48
+ "train": "https://drive.google.com/uc?export=download&id=1S3vNx9By5CWKGkObjtXaI6Jr4xri2Tz3",
49
+ "valid": "https://drive.google.com/uc?export=download&id=1cP-zDIxp9a-Bw9uYd40K9IN-4wg4dOgy",
50
+ "test": "https://drive.google.com/uc?export=download&id=1pLcoJgYmgQiN4Gv9tRcI26zM7-OgHcuZ",
51
+ },
52
+ "transductive": {
53
+ "train": "https://drive.google.com/uc?export=download&id=1KXDVwboo1h2yk_kAqv7IPYnHXCK6g-6X",
54
+ "valid": "https://drive.google.com/uc?export=download&id=1eRwpuRPYOnA-7FZ-YNZjRJ2DHuJsfUIE",
55
+ "test": "https://drive.google.com/uc?export=download&id=1cy9FwDMB_U-js8P8u4IWolvNeIFkQVDh",
56
+ },
57
+ "text": "https://drive.usercontent.google.com/download?id=1YC4P_IPSo1AsEwm5Z_4GBjDdwCbvokxX&export=download&authuser=0&confirm=t&uuid=36aa95f5-e1b6-43c1-a34f-754d14d8b473&at=APZUnTWD7fwarBs4ZVRy_QdKbDXi%3A1709478240158",
58
+ }
59
+
60
+ # none of the tasks in schema
61
+ # dataset is used to learn knowledge embedding
62
+ _SUPPORTED_TASKS = []
63
+
64
+ _SOURCE_VERSION = "1.0.0"
65
+ _SEACROWD_VERSION = "2024.06.20"
66
+
67
+
68
+ class IndoWiki(datasets.GeneratorBasedBuilder):
69
+ """IndoWiki knowledge base dataset from https://github.com/IgoRamli/IndoWiki"""
70
+
71
+ SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
72
+ SEACROWD_VERSION = datasets.Version(_SEACROWD_VERSION)
73
+
74
+ BUILDER_CONFIGS = [
75
+ # inductive setting
76
+ SEACrowdConfig(
77
+ name=f"{_DATASETNAME}_inductive_source",
78
+ version=SOURCE_VERSION,
79
+ description=f"{_DATASETNAME} source schema",
80
+ schema="source",
81
+ subset_id=_DATASETNAME,
82
+ ),
83
+ # transductive setting
84
+ SEACrowdConfig(
85
+ name=f"{_DATASETNAME}_source",
86
+ version=SOURCE_VERSION,
87
+ description=f"{_DATASETNAME} source schema",
88
+ schema="source",
89
+ subset_id=_DATASETNAME,
90
+ ),
91
+ ]
92
+
93
+ DEFAULT_CONFIG_NAME = f"{_DATASETNAME}_source"
94
+
95
+ def _info(self) -> datasets.DatasetInfo:
96
+
97
+ if self.config.schema == "source":
98
+ features = datasets.Features(
99
+ {
100
+ "id": datasets.Value("string"),
101
+ "ent1": datasets.Value("string"),
102
+ "ent2": datasets.Value("string"),
103
+ "ent1_text": datasets.Value("string"),
104
+ "ent2_text": datasets.Value("string"),
105
+ "relation": datasets.Value("string"),
106
+ }
107
+ )
108
+
109
+ else:
110
+ raise NotImplementedError()
111
+
112
+ return datasets.DatasetInfo(
113
+ description=_DESCRIPTION,
114
+ features=features,
115
+ homepage=_HOMEPAGE,
116
+ license=_LICENSE,
117
+ citation=_CITATION,
118
+ )
119
+
120
+ def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
121
+ """Returns SplitGenerators."""
122
+
123
+ if "inductive" in self.config.name:
124
+ setting = "inductive"
125
+ data_paths = {
126
+ "inductive": {
127
+ "train": Path(dl_manager.download_and_extract(_URLS["inductive"]["train"])),
128
+ "valid": Path(dl_manager.download_and_extract(_URLS["inductive"]["valid"])),
129
+ "test": Path(dl_manager.download_and_extract(_URLS["inductive"]["test"])),
130
+ },
131
+ "text": Path(dl_manager.download_and_extract(_URLS["text"])),
132
+ }
133
+ else:
134
+ setting = "transductive"
135
+ data_paths = {
136
+ "transductive": {
137
+ "train": Path(dl_manager.download_and_extract(_URLS["transductive"]["train"])),
138
+ "valid": Path(dl_manager.download_and_extract(_URLS["transductive"]["valid"])),
139
+ "test": Path(dl_manager.download_and_extract(_URLS["transductive"]["test"])),
140
+ },
141
+ "text": Path(dl_manager.download_and_extract(_URLS["text"])),
142
+ }
143
+
144
+ return [
145
+ datasets.SplitGenerator(
146
+ name=datasets.Split.TRAIN,
147
+ gen_kwargs={
148
+ "triplets_filepath": data_paths[setting]["train"],
149
+ "text_filepath": data_paths["text"],
150
+ "split": "train",
151
+ },
152
+ ),
153
+ datasets.SplitGenerator(
154
+ name=datasets.Split.TEST,
155
+ gen_kwargs={
156
+ "triplets_filepath": data_paths[setting]["test"],
157
+ "text_filepath": data_paths["text"],
158
+ "split": "test",
159
+ },
160
+ ),
161
+ datasets.SplitGenerator(
162
+ name=datasets.Split.VALIDATION,
163
+ gen_kwargs={
164
+ "triplets_filepath": data_paths[setting]["valid"],
165
+ "text_filepath": data_paths["text"],
166
+ "split": "dev",
167
+ },
168
+ ),
169
+ ]
170
+
171
+ def _generate_examples(self, triplets_filepath: Path, text_filepath: Path, split: str) -> Tuple[int, Dict]:
172
+ """Yields examples as (key, example) tuples."""
173
+
174
+ # read triplets file
175
+ with open(triplets_filepath, "r", encoding="utf-8") as triplets_file:
176
+ triplets_data = triplets_file.readlines()
177
+ triplets_data = [s.strip("\n").split("\t") for s in triplets_data]
178
+
179
+ # read text description file
180
+ with open(text_filepath, "r", encoding="utf-8") as text_file:
181
+ text_data = text_file.readlines()
182
+ # dictionary of entity: text description of entity
183
+ text_dict = {s.split("\t")[0]: s.split("\t")[1].strip("\n") for s in text_data}
184
+
185
+ num_sample = len(triplets_data)
186
+
187
+ for i in range(num_sample):
188
+ if self.config.schema == "source":
189
+ example = {
190
+ "id": str(i),
191
+ "ent1": triplets_data[i][0],
192
+ "ent2": triplets_data[i][2],
193
+ "ent1_text": text_dict[triplets_data[i][0]],
194
+ "ent2_text": text_dict[triplets_data[i][2]],
195
+ "relation": triplets_data[i][1],
196
+ }
197
+
198
+ yield i, example