Upload 2 files
Browse files
KB.py
ADDED
@@ -0,0 +1,103 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import json
|
2 |
+
import math
|
3 |
+
from typing import List
|
4 |
+
|
5 |
+
import datasets
|
6 |
+
|
7 |
+
|
8 |
+
logger = datasets.logging.get_logger(__name__)
|
9 |
+
|
10 |
+
|
11 |
+
_CITATION = ''
|
12 |
+
|
13 |
+
_DESCRIPTION = """\
|
14 |
+
Выдержки из руководства пользователя для генерации ответов по запросу пользователя.\
|
15 |
+
"""
|
16 |
+
_LICENSE = ''
|
17 |
+
|
18 |
+
_URL = "https://huggingface.co/datasets/SIA86/KB/blob/main/"
|
19 |
+
_URLS = {
|
20 |
+
"train": _URL + "kb.json",
|
21 |
+
}
|
22 |
+
|
23 |
+
class LFQAKBConfig(datasets.BuilderConfig):
|
24 |
+
"""BuilderConfig for LFQAKnowledgeBase"""
|
25 |
+
|
26 |
+
def __init__(
|
27 |
+
self, language="ru", snippets_length=50, snippets_overlap=25, **kwargs
|
28 |
+
):
|
29 |
+
"""BuilderConfig for WikiSnippets.
|
30 |
+
Args:
|
31 |
+
**kwargs: keyword arguments forwarded to super.
|
32 |
+
"""
|
33 |
+
super(LFQAKBConfig, self).__init__(**kwargs)
|
34 |
+
self.language = language
|
35 |
+
self.snippets_length = snippets_length
|
36 |
+
self.snippets_overlap = snippets_overlap
|
37 |
+
|
38 |
+
|
39 |
+
class WikiSnippets(datasets.GeneratorBasedBuilder):
|
40 |
+
|
41 |
+
BUILDER_CONFIGS = LFQAKBConfig(
|
42 |
+
name="LFQAKnowledgeBase",
|
43 |
+
version=datasets.Version("1.0.0"),
|
44 |
+
language="en",
|
45 |
+
snippets_length=50,
|
46 |
+
snippets_overlap=25,
|
47 |
+
)
|
48 |
+
|
49 |
+
test_dummy_data = False
|
50 |
+
|
51 |
+
def _info(self):
|
52 |
+
return datasets.DatasetInfo(
|
53 |
+
description=_DESCRIPTION,
|
54 |
+
features=datasets.Features(
|
55 |
+
{
|
56 |
+
"id": datasets.Value("int32g"),
|
57 |
+
"title": datasets.Value("string"),
|
58 |
+
"heading1": datasets.Value("string"),
|
59 |
+
"heading2": datasets.Value("string"),
|
60 |
+
"heading3": datasets.Value("string"),
|
61 |
+
"heading4": datasets.Value("string"),
|
62 |
+
"heading5": datasets.Value("string"),
|
63 |
+
"text": datasets.Value("string"),
|
64 |
+
|
65 |
+
}
|
66 |
+
),
|
67 |
+
supervised_keys=None,
|
68 |
+
citation=_CITATION,
|
69 |
+
)
|
70 |
+
|
71 |
+
def _split_generators(self, dl_manager: datasets.DownloadManager) -> List[datasets.SplitGenerator]:
|
72 |
+
urls_to_download = self._URLS
|
73 |
+
downloaded_files = dl_manager.download_and_extract(urls_to_download)
|
74 |
+
|
75 |
+
return [
|
76 |
+
datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files}),
|
77 |
+
]
|
78 |
+
|
79 |
+
def _generate_examples(self, filepath):
|
80 |
+
"""This function returns the examples in the raw (text) form."""
|
81 |
+
logger.info("generating examples from = %s", filepath)
|
82 |
+
with open(filepath) as f:
|
83 |
+
lfqa = json.load(f)
|
84 |
+
for article in lfqa:
|
85 |
+
id_ = article['id']
|
86 |
+
title = article['title']
|
87 |
+
heading1 = article['heading1']
|
88 |
+
heading2 = article['heading2']
|
89 |
+
heading3 = article['heading3']
|
90 |
+
heading4 = article['heading4']
|
91 |
+
heading5 = article['heading5']
|
92 |
+
text = article['text']
|
93 |
+
# Features currently used are "context", "question", and "answers".
|
94 |
+
# Others are extracted here for the ease of future expansions.
|
95 |
+
yield id_, {
|
96 |
+
"title": title,
|
97 |
+
"heading1": heading1,
|
98 |
+
"heading2": heading2,
|
99 |
+
"heading3": heading3,
|
100 |
+
"heading4": heading4,
|
101 |
+
"heading5": heading5,
|
102 |
+
"text": text
|
103 |
+
}
|
kb.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|