gabrielaltay
commited on
Commit
·
90953a7
1
Parent(s):
23c07e4
upload hubscripts/chebi_nactem_hub.py to hub from bigbio repo
Browse files- chebi_nactem.py +237 -0
chebi_nactem.py
ADDED
@@ -0,0 +1,237 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
import os
|
17 |
+
from pathlib import Path
|
18 |
+
from typing import Dict, List, Tuple
|
19 |
+
|
20 |
+
import datasets
|
21 |
+
|
22 |
+
from .bigbiohub import kb_features
|
23 |
+
from .bigbiohub import BigBioConfig
|
24 |
+
from .bigbiohub import Tasks
|
25 |
+
|
26 |
+
_LANGUAGES = ['English']
|
27 |
+
_PUBMED = True
|
28 |
+
_LOCAL = False
|
29 |
+
_CITATION = """\
|
30 |
+
@inproceedings{Shardlow2018,
|
31 |
+
title = {
|
32 |
+
A New Corpus to Support Text Mining for the Curation of Metabolites in the
|
33 |
+
{ChEBI} Database
|
34 |
+
},
|
35 |
+
author = {
|
36 |
+
Shardlow, M J and Nguyen, N and Owen, G and O'Donovan, C and Leach, A and
|
37 |
+
McNaught, J and Turner, S and Ananiadou, S
|
38 |
+
},
|
39 |
+
year = 2018,
|
40 |
+
month = may,
|
41 |
+
booktitle = {
|
42 |
+
Proceedings of the Eleventh International Conference on Language Resources
|
43 |
+
and Evaluation ({LREC} 2018)
|
44 |
+
},
|
45 |
+
location = {Miyazaki, Japan},
|
46 |
+
pages = {280--285},
|
47 |
+
conference = {
|
48 |
+
Eleventh International Conference on Language Resources and Evaluation
|
49 |
+
(LREC 2018)
|
50 |
+
},
|
51 |
+
language = {en}
|
52 |
+
}
|
53 |
+
"""
|
54 |
+
|
55 |
+
_DATASETNAME = "chebi_nactem"
|
56 |
+
_DISPLAYNAME = "CHEBI Corpus"
|
57 |
+
|
58 |
+
_DESCRIPTION = """\
|
59 |
+
The ChEBI corpus contains 199 annotated abstracts and 100 annotated full papers.
|
60 |
+
All documents in the corpus have been annotated for named entities and relations
|
61 |
+
between these. In total, our corpus provides over 15000 named entity annotations
|
62 |
+
and over 6,000 relations between entities.
|
63 |
+
"""
|
64 |
+
|
65 |
+
_HOMEPAGE = "http://www.nactem.ac.uk/chebi"
|
66 |
+
|
67 |
+
_LICENSE = 'Creative Commons Attribution 4.0 International'
|
68 |
+
|
69 |
+
_URLS = {
|
70 |
+
_DATASETNAME: "http://www.nactem.ac.uk/chebi/ChEBI.zip",
|
71 |
+
}
|
72 |
+
|
73 |
+
_SUPPORTED_TASKS = [Tasks.NAMED_ENTITY_RECOGNITION, Tasks.RELATION_EXTRACTION]
|
74 |
+
|
75 |
+
_SOURCE_VERSION = "1.0.0"
|
76 |
+
|
77 |
+
_BIGBIO_VERSION = "1.0.0"
|
78 |
+
|
79 |
+
|
80 |
+
class ChebiNactemDatasset(datasets.GeneratorBasedBuilder):
|
81 |
+
"""Chemical Entities of Biological Interest (ChEBI) corpus."""
|
82 |
+
|
83 |
+
SOURCE_VERSION = datasets.Version(_SOURCE_VERSION)
|
84 |
+
BIGBIO_VERSION = datasets.Version(_BIGBIO_VERSION)
|
85 |
+
|
86 |
+
BUILDER_CONFIGS = []
|
87 |
+
for subset_id in ["abstr_ann1", "abstr_ann2", "fullpaper"]:
|
88 |
+
BUILDER_CONFIGS += [
|
89 |
+
BigBioConfig(
|
90 |
+
name=f"chebi_nactem_{subset_id}_source",
|
91 |
+
version=SOURCE_VERSION,
|
92 |
+
description="chebi_nactem source schema",
|
93 |
+
schema="source",
|
94 |
+
subset_id=f"chebi_nactem_{subset_id}",
|
95 |
+
),
|
96 |
+
BigBioConfig(
|
97 |
+
name=f"chebi_nactem_{subset_id}_bigbio_kb",
|
98 |
+
version=BIGBIO_VERSION,
|
99 |
+
description="chebi_nactem BigBio schema",
|
100 |
+
schema="bigbio_kb",
|
101 |
+
subset_id=f"chebi_nactem_{subset_id}",
|
102 |
+
),
|
103 |
+
]
|
104 |
+
|
105 |
+
DEFAULT_CONFIG_NAME = "chebi_nactem_fullpaper_source"
|
106 |
+
|
107 |
+
def _info(self) -> datasets.DatasetInfo:
|
108 |
+
if self.config.schema == "source":
|
109 |
+
features = datasets.Features(
|
110 |
+
{
|
111 |
+
"document_id": datasets.Value("string"),
|
112 |
+
"text": datasets.Value("string"),
|
113 |
+
"entities": [
|
114 |
+
{
|
115 |
+
"id": datasets.Value("string"),
|
116 |
+
"type": datasets.Value("string"),
|
117 |
+
"text": datasets.Value("string"),
|
118 |
+
"offsets": datasets.Sequence([datasets.Value("int32")]),
|
119 |
+
}
|
120 |
+
],
|
121 |
+
"relations": [
|
122 |
+
{
|
123 |
+
"id": datasets.Value("string"),
|
124 |
+
"type": datasets.Value("string"),
|
125 |
+
"arg1": datasets.Value("string"),
|
126 |
+
"arg2": datasets.Value("string"),
|
127 |
+
}
|
128 |
+
],
|
129 |
+
}
|
130 |
+
)
|
131 |
+
|
132 |
+
elif self.config.schema == "bigbio_kb":
|
133 |
+
features = kb_features
|
134 |
+
else:
|
135 |
+
raise NotImplementedError(self.config.schema)
|
136 |
+
|
137 |
+
return datasets.DatasetInfo(
|
138 |
+
description=_DESCRIPTION,
|
139 |
+
features=features,
|
140 |
+
homepage=_HOMEPAGE,
|
141 |
+
license=str(_LICENSE),
|
142 |
+
citation=_CITATION,
|
143 |
+
)
|
144 |
+
|
145 |
+
def _split_generators(self, dl_manager) -> List[datasets.SplitGenerator]:
|
146 |
+
"""Returns SplitGenerators."""
|
147 |
+
urls = _URLS[_DATASETNAME]
|
148 |
+
data_dir = dl_manager.download_and_extract(urls)
|
149 |
+
|
150 |
+
subset_paths = {
|
151 |
+
"chebi_nactem_abstr_ann1": "ChEBI/abstracts/Annotator1",
|
152 |
+
"chebi_nactem_abstr_ann2": "ChEBI/abstracts/Annotator2",
|
153 |
+
"chebi_nactem_fullpaper": "ChEBI/fullpapers",
|
154 |
+
}
|
155 |
+
|
156 |
+
subset_dir = Path(data_dir) / subset_paths[self.config.subset_id]
|
157 |
+
|
158 |
+
return [
|
159 |
+
datasets.SplitGenerator(
|
160 |
+
name=datasets.Split.TRAIN,
|
161 |
+
# Whatever you put in gen_kwargs will be passed to _generate_examples
|
162 |
+
gen_kwargs={
|
163 |
+
"data_dir": subset_dir,
|
164 |
+
},
|
165 |
+
)
|
166 |
+
]
|
167 |
+
|
168 |
+
def _generate_examples(self, data_dir: Path) -> Tuple[int, Dict]:
|
169 |
+
"""Yields examples as (key, example) tuples."""
|
170 |
+
|
171 |
+
def uid_gen():
|
172 |
+
_uid = 0
|
173 |
+
while True:
|
174 |
+
yield str(_uid)
|
175 |
+
_uid += 1
|
176 |
+
|
177 |
+
uid = iter(uid_gen())
|
178 |
+
|
179 |
+
txt_files = (f for f in os.listdir(data_dir) if f.endswith(".txt"))
|
180 |
+
for idx, file_name in enumerate(txt_files):
|
181 |
+
|
182 |
+
brat_file = data_dir / file_name
|
183 |
+
contents = parse_brat_file(brat_file)
|
184 |
+
|
185 |
+
if self.config.schema == "source":
|
186 |
+
yield idx, {
|
187 |
+
"document_id": contents["document_id"],
|
188 |
+
"text": contents["text"],
|
189 |
+
"entities": contents["text_bound_annotations"],
|
190 |
+
"relations": [
|
191 |
+
{
|
192 |
+
"id": relation["id"],
|
193 |
+
"type": relation["type"],
|
194 |
+
"arg1": relation["head"]["ref_id"],
|
195 |
+
"arg2": relation["tail"]["ref_id"],
|
196 |
+
}
|
197 |
+
for relation in contents["relations"]
|
198 |
+
],
|
199 |
+
}
|
200 |
+
|
201 |
+
elif self.config.schema == "bigbio_kb":
|
202 |
+
yield idx, {
|
203 |
+
"id": next(uid),
|
204 |
+
"document_id": contents["document_id"],
|
205 |
+
"passages": [
|
206 |
+
{
|
207 |
+
"id": next(uid),
|
208 |
+
"type": "",
|
209 |
+
"text": [contents["text"]],
|
210 |
+
"offsets": [(0, len(contents["text"]))],
|
211 |
+
}
|
212 |
+
],
|
213 |
+
"entities": [
|
214 |
+
{
|
215 |
+
"id": f"{idx}_{entity['id']}",
|
216 |
+
"type": entity["type"],
|
217 |
+
"offsets": entity["offsets"],
|
218 |
+
"text": entity["text"],
|
219 |
+
"normalized": [],
|
220 |
+
}
|
221 |
+
for entity in contents["text_bound_annotations"]
|
222 |
+
],
|
223 |
+
"events": [],
|
224 |
+
"coreferences": [],
|
225 |
+
"relations": [
|
226 |
+
{
|
227 |
+
"id": f"{idx}_{relation['id']}",
|
228 |
+
"type": relation["type"],
|
229 |
+
"arg1_id": f"{idx}_{relation['head']['ref_id']}",
|
230 |
+
"arg2_id": f"{idx}_{relation['tail']['ref_id']}",
|
231 |
+
"normalized": [],
|
232 |
+
}
|
233 |
+
for relation in contents["relations"]
|
234 |
+
],
|
235 |
+
}
|
236 |
+
else:
|
237 |
+
raise NotImplementedError(self.config.schema)
|