zeio commited on
Commit
219741e
·
verified ·
1 Parent(s): 36c56da

feat(dataset-loader): added dataset loader script

Browse files
Files changed (1) hide show
  1. pale.py +83 -0
pale.py ADDED
@@ -0,0 +1,83 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from pandas import read_csv
2
+
3
+ from datasets import GeneratorBasedBuilder, Value, Version, BuilderConfig, Features, DatasetInfo, SplitGenerator, Split
4
+
5
+ _DESCRIPTION = '''
6
+ This dataset contains transcribed sounds emitted by characters of the League of Legends game.
7
+ The data can be useful for building text classification models, fine-tuning language generation models, speech synthesis and speech recognition models.
8
+ The underlying web dump for the dataset construction has been last refreshed **20.10.2023**
9
+ '''
10
+
11
+ _HOMEPAGE = 'https://huggingface.co/datasets/zeio/pale'
12
+
13
+ _LICENSE = 'Apache License Version 2.0'
14
+
15
+ _URLS = {
16
+ 'full': 'https://huggingface.co/datasets/zeio/pale/resolve/main/pale.tsv',
17
+ 'quotes': 'https://huggingface.co/datasets/zeio/pale/resolve/main/quotes.tsv',
18
+ 'annotated': 'https://huggingface.co/datasets/zeio/pale/resolve/main/annotated.tsv'
19
+ }
20
+
21
+
22
+ class Pale(GeneratorBasedBuilder):
23
+
24
+ VERSION = Version('27.10.2023')
25
+
26
+ BUILDER_CONFIGS = [
27
+ BuilderConfig(name = 'quotes', version = VERSION, description = 'Truncated version of the corpus, which does\'t contain sound effects'),
28
+ BuilderConfig(name = 'annotated', version = VERSION, description = 'An extended version of the full configuration with a couple of additional columns with labels'),
29
+ BuilderConfig(name = 'vanilla', version = VERSION, description = 'All data pulled from the website without significant modifications apart from the web page structure parsing')
30
+ ]
31
+
32
+ DEFAULT_CONFIG_NAME = 'quotes'
33
+
34
+ def _info(self):
35
+ if self.config.name == 'vanilla':
36
+ features = Features({
37
+ 'header': Value('string'),
38
+ 'subheader': Value('string'),
39
+ 'text': Value('string'),
40
+ 'source': Value('string'),
41
+ 'champion': Value('string')
42
+ })
43
+ elif self.config.name == 'annotated':
44
+ features = Features({
45
+ 'header': Value('string'),
46
+ 'subheader': Value('string'),
47
+ 'text': Value('string'),
48
+ 'source': Value('string'),
49
+ 'champion': Value('string'),
50
+ 'quote': Value('bool')
51
+ })
52
+ elif self.config.name == 'quotes':
53
+ features = Features({
54
+ 'header': Value('string'),
55
+ 'subheader': Value('string'),
56
+ 'text': Value('string'),
57
+ 'champion': Value('string')
58
+ })
59
+
60
+ return DatasetInfo(
61
+ description=_DESCRIPTION,
62
+ features = features,
63
+ homepage=_HOMEPAGE,
64
+ license=_LICENSE
65
+ )
66
+
67
+ def _split_generators(self, dl_manager):
68
+ name = self.config.name
69
+
70
+ url = _URLS[name]
71
+
72
+ return [
73
+ SplitGenerator(
74
+ name = Split.TRAIN,
75
+ gen_kwargs = {
76
+ "path": dl_manager.download_and_extract(url)
77
+ }
78
+ )
79
+ ]
80
+
81
+ def _generate_examples(self, path: str):
82
+ for i, row in read_csv(path, sep = '\t').iterrows():
83
+ yield i, dict(row)