Commit
·
7d04c1d
0
Parent(s):
v1.0.0
Browse files- .gitattributes +38 -0
- README.md +65 -0
- RuREBus.py +142 -0
- data/test.jsonl +0 -0
- data/train.jsonl +0 -0
.gitattributes
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
25 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
26 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
27 |
+
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
28 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
29 |
+
# Audio files - uncompressed
|
30 |
+
*.pcm filter=lfs diff=lfs merge=lfs -text
|
31 |
+
*.sam filter=lfs diff=lfs merge=lfs -text
|
32 |
+
*.raw filter=lfs diff=lfs merge=lfs -text
|
33 |
+
# Audio files - compressed
|
34 |
+
*.aac filter=lfs diff=lfs merge=lfs -text
|
35 |
+
*.flac filter=lfs diff=lfs merge=lfs -text
|
36 |
+
*.mp3 filter=lfs diff=lfs merge=lfs -text
|
37 |
+
*.ogg filter=lfs diff=lfs merge=lfs -text
|
38 |
+
*.wav filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
languages:
|
3 |
+
- ru
|
4 |
+
multilinguality:
|
5 |
+
- monolingual
|
6 |
+
pretty_name: RuREBus
|
7 |
+
task_categories:
|
8 |
+
- structure-prediction
|
9 |
+
task_ids:
|
10 |
+
- named-entity-recognition
|
11 |
+
---
|
12 |
+
|
13 |
+
# RuREBus dataset
|
14 |
+
|
15 |
+
## Table of Contents
|
16 |
+
- [Dataset Description](#dataset-description)
|
17 |
+
- [Dataset Structure](#dataset-structure)
|
18 |
+
- [Citation Information](#citation-information)
|
19 |
+
- [Contacts](#contacts)
|
20 |
+
|
21 |
+
## Dataset Description
|
22 |
+
RuREBus dataset (https://github.com/dialogue-evaluation/RuREBus) is
|
23 |
+
a Russian dataset for named entity recognition and relation extraction.
|
24 |
+
|
25 |
+
## Dataset Structure
|
26 |
+
There are two subsets of the dataset.
|
27 |
+
|
28 |
+
Using
|
29 |
+
`load_dataset('MalakhovIlya/RuREBus')`
|
30 |
+
you can download annotated data (DatasetDict) for named entity recognition task and
|
31 |
+
relation extraction tasks.
|
32 |
+
This subset consists of two splits: "train" and "test".
|
33 |
+
|
34 |
+
Using
|
35 |
+
`load_dataset('MalakhovIlya/NEREL', 'raw_txt')['raw_txt']`
|
36 |
+
you can download (Dataset) large corpus (~3gb) raw texts of the same subject
|
37 |
+
area, but without any annotations.
|
38 |
+
|
39 |
+
"entities" are used in named-entity recognition task (see https://en.wikipedia.org/wiki/Named-entity_recognition).
|
40 |
+
"relations" are used in relationship extraction task (see https://en.wikipedia.org/wiki/Relationship_extraction).
|
41 |
+
|
42 |
+
Each entity is represented by a string of the following format:
|
43 |
+
`"<id>\t<type> <start> <stop>\t<text>"`, where
|
44 |
+
`<id>` is an entity id,
|
45 |
+
`<type>` is one of entity types,
|
46 |
+
`<start>` is a position of the first symbol of entity in text,
|
47 |
+
`<stop>` is the last symbol position in text +1.
|
48 |
+
|
49 |
+
Each relation is represented by a string of the following format:
|
50 |
+
`"<id>\t<type> Arg1:<arg1_id> Arg2:<arg2_id>"`, where
|
51 |
+
`<id>` is a relation id,
|
52 |
+
`<arg1_id>` and `<arg2_id>` are entity ids.
|
53 |
+
|
54 |
+
## Citation Information
|
55 |
+
@inproceedings{rurebus,
|
56 |
+
Address = {Moscow, Russia},
|
57 |
+
Author = {Ivanin, Vitaly and Artemova, Ekaterina and Batura, Tatiana and Ivanov, Vladimir and Sarkisyan, Veronika and Tutubalina, Elena and Smurov, Ivan},
|
58 |
+
Title = {RuREBus-2020 Shared Task: Russian Relation Extraction for Business},
|
59 |
+
Booktitle = {Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialog” [Komp’iuternaia Lingvistika i Intellektual’nye Tehnologii: Trudy Mezhdunarodnoj Konferentsii “Dialog”]},
|
60 |
+
Year = {2020}
|
61 |
+
}
|
62 |
+
|
63 |
+
## Contacts
|
64 |
+
Malakhov Ilya
|
65 |
+
Telegram - https://t.me/noname_4710
|
RuREBus.py
ADDED
@@ -0,0 +1,142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import datasets
|
2 |
+
import json
|
3 |
+
import requests
|
4 |
+
from urllib.parse import urlencode
|
5 |
+
from pathlib import Path
|
6 |
+
import zipfile
|
7 |
+
import os
|
8 |
+
import shutil
|
9 |
+
|
10 |
+
_NAME = 'RuREBus'
|
11 |
+
_CITATION = '''
|
12 |
+
@inproceedings{rurebus,
|
13 |
+
Address = {Moscow, Russia},
|
14 |
+
Author = {Ivanin, Vitaly and Artemova, Ekaterina and Batura, Tatiana and Ivanov, Vladimir and Sarkisyan, Veronika and Tutubalina, Elena and Smurov, Ivan},
|
15 |
+
Title = {RuREBus-2020 Shared Task: Russian Relation Extraction for Business},
|
16 |
+
Booktitle = {Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialog” [Komp’iuternaia Lingvistika i Intellektual’nye Tehnologii: Trudy Mezhdunarodnoj Konferentsii “Dialog”]},
|
17 |
+
Year = {2020}
|
18 |
+
}
|
19 |
+
'''.strip()
|
20 |
+
_DESCRIPTION = 'Russian Relation Extraction for Business'
|
21 |
+
_HOMEPAGE = 'https://github.com/dialogue-evaluation/RuREBus'
|
22 |
+
_VERSION = '1.0.0'
|
23 |
+
|
24 |
+
|
25 |
+
def decode_file_names(folder):
|
26 |
+
for root, dirs, files in os.walk(folder, topdown=False):
|
27 |
+
root = Path(root)
|
28 |
+
for file in files:
|
29 |
+
old_name = root / Path(file)
|
30 |
+
new_name = root / Path(
|
31 |
+
file.encode('cp437').decode('cp866'))
|
32 |
+
old_name.rename(new_name)
|
33 |
+
for dir in dirs:
|
34 |
+
old_name = root / Path(dir)
|
35 |
+
new_name = root / Path(dir.encode('cp437').decode('cp866'))
|
36 |
+
old_name.rename(new_name)
|
37 |
+
|
38 |
+
|
39 |
+
class RuREBusBuilder(datasets.GeneratorBasedBuilder):
|
40 |
+
base_url = 'https://cloud-api.yandex.net/v1/disk/public/resources/download?'
|
41 |
+
public_key = 'https://disk.yandex.ru/d/t1WakmYXlL6jBw'
|
42 |
+
final_url = base_url + urlencode(dict(public_key=public_key))
|
43 |
+
response = requests.get(final_url)
|
44 |
+
raw_txt_url = response.json()['href']
|
45 |
+
|
46 |
+
_DATA_URLS = {
|
47 |
+
'train': 'data/train.jsonl',
|
48 |
+
'test': f'data/test.jsonl',
|
49 |
+
}
|
50 |
+
_RAW_TXT_URLS = {
|
51 |
+
'raw_txt': raw_txt_url
|
52 |
+
}
|
53 |
+
VERSION = datasets.Version(_VERSION)
|
54 |
+
BUILDER_CONFIGS = [
|
55 |
+
datasets.BuilderConfig('data',
|
56 |
+
version=VERSION,
|
57 |
+
description='Annotated data'),
|
58 |
+
datasets.BuilderConfig('raw_txt',
|
59 |
+
version=VERSION,
|
60 |
+
description='Raw texts without annotations'),
|
61 |
+
]
|
62 |
+
DEFAULT_CONFIG_NAME = 'data'
|
63 |
+
|
64 |
+
def _info(self) -> datasets.DatasetInfo:
|
65 |
+
if self.config.name == 'data':
|
66 |
+
features = datasets.Features({
|
67 |
+
'id': datasets.Value('int32'),
|
68 |
+
'text': datasets.Value('string'),
|
69 |
+
'entities': datasets.Sequence(datasets.Value('string')),
|
70 |
+
'relations': datasets.Sequence(datasets.Value('string'))
|
71 |
+
})
|
72 |
+
else:
|
73 |
+
features = datasets.Features({
|
74 |
+
'region': datasets.Value('string'),
|
75 |
+
'district': datasets.Value('string'),
|
76 |
+
'title': datasets.Value('string'),
|
77 |
+
'text': datasets.Value('string')
|
78 |
+
})
|
79 |
+
return datasets.DatasetInfo(
|
80 |
+
description=_DESCRIPTION,
|
81 |
+
features=features,
|
82 |
+
homepage=_HOMEPAGE,
|
83 |
+
citation=_CITATION
|
84 |
+
)
|
85 |
+
|
86 |
+
def _split_generators(self, dl_manager: datasets.DownloadManager):
|
87 |
+
if self.config.name == 'data':
|
88 |
+
files = dl_manager.download(self._DATA_URLS)
|
89 |
+
return [
|
90 |
+
datasets.SplitGenerator(
|
91 |
+
name=datasets.Split.TRAIN,
|
92 |
+
gen_kwargs={'filepath': files['train']},
|
93 |
+
),
|
94 |
+
datasets.SplitGenerator(
|
95 |
+
name=datasets.Split.TEST,
|
96 |
+
gen_kwargs={'filepath': files['test']},
|
97 |
+
),
|
98 |
+
]
|
99 |
+
else:
|
100 |
+
folder = dl_manager.download_and_extract(self._RAW_TXT_URLS)['raw_txt']
|
101 |
+
decode_file_names(folder)
|
102 |
+
return [
|
103 |
+
datasets.SplitGenerator(
|
104 |
+
name='raw_txt',
|
105 |
+
gen_kwargs={'filepath': folder,
|
106 |
+
'raw_txt': True},
|
107 |
+
)
|
108 |
+
]
|
109 |
+
|
110 |
+
def _generate_examples(self, filepath, raw_txt=False):
|
111 |
+
if not raw_txt:
|
112 |
+
with open(filepath, encoding='utf-8') as f:
|
113 |
+
for i, line in enumerate(f):
|
114 |
+
yield i, json.loads(line)
|
115 |
+
else:
|
116 |
+
path = Path(filepath) / 'MED_txt/unparsed_txt'
|
117 |
+
i = 0
|
118 |
+
for root, dirs, files in os.walk(path):
|
119 |
+
if files:
|
120 |
+
root = Path(root)
|
121 |
+
region = root.parent.name
|
122 |
+
district = root.name
|
123 |
+
titles = {}
|
124 |
+
with open(root / 'name_dict.txt', encoding='utf-8') as f_titles:
|
125 |
+
for line in f_titles:
|
126 |
+
key, title = line.split(maxsplit=1)[1].split('_', maxsplit=1)
|
127 |
+
titles[key] = title.strip()
|
128 |
+
for file in files:
|
129 |
+
if file != 'name_dict.txt':
|
130 |
+
file = Path(file)
|
131 |
+
key = file.name.split('_', maxsplit=1)[0]
|
132 |
+
title = titles[key]
|
133 |
+
with open(root / file, encoding='utf-8') as f:
|
134 |
+
text = f.read()
|
135 |
+
item = {
|
136 |
+
'region': region,
|
137 |
+
'district': district,
|
138 |
+
'title': title,
|
139 |
+
'text': text
|
140 |
+
}
|
141 |
+
yield i, item
|
142 |
+
i += 1
|
data/test.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|
data/train.jsonl
ADDED
The diff for this file is too large to render.
See raw diff
|
|