Datasets:
feat(pulled): added support for pulled dataset config
Browse files
pale.py
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
-
|
|
|
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.
|
@@ -15,9 +16,12 @@ _LICENSE = 'Apache License Version 2.0'
|
|
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 |
|
@@ -26,7 +30,8 @@ class Pale(GeneratorBasedBuilder):
|
|
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'
|
@@ -56,6 +61,16 @@ class Pale(GeneratorBasedBuilder):
|
|
56 |
'text': Value('string'),
|
57 |
'champion': Value('string')
|
58 |
})
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
59 |
|
60 |
return DatasetInfo(
|
61 |
description=_DESCRIPTION,
|
@@ -73,11 +88,28 @@ class Pale(GeneratorBasedBuilder):
|
|
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 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
import os
|
2 |
+
from pandas import read_csv, NA
|
3 |
|
4 |
+
from datasets import GeneratorBasedBuilder, Value, Version, BuilderConfig, Features, DatasetInfo, SplitGenerator, Split, Audio
|
5 |
|
6 |
_DESCRIPTION = '''
|
7 |
This dataset contains transcribed sounds emitted by characters of the League of Legends game.
|
|
|
16 |
_URLS = {
|
17 |
'full': 'https://huggingface.co/datasets/zeio/pale/resolve/main/pale.tsv',
|
18 |
'quotes': 'https://huggingface.co/datasets/zeio/pale/resolve/main/quotes.tsv',
|
19 |
+
'annotated': 'https://huggingface.co/datasets/zeio/pale/resolve/main/annotated.tsv',
|
20 |
+
'pulled': 'https://huggingface.co/datasets/zeio/pale/resolve/main/pulled.tsv'
|
21 |
}
|
22 |
|
23 |
+
_SOUND_URL = 'https://huggingface.co/datasets/zeio/pale/resolve/main/sound.tar.xz'
|
24 |
+
|
25 |
|
26 |
class Pale(GeneratorBasedBuilder):
|
27 |
|
|
|
30 |
BUILDER_CONFIGS = [
|
31 |
BuilderConfig(name = 'quotes', version = VERSION, description = 'Truncated version of the corpus, which does\'t contain sound effects'),
|
32 |
BuilderConfig(name = 'annotated', version = VERSION, description = 'An extended version of the full configuration with a couple of additional columns with labels'),
|
33 |
+
BuilderConfig(name = 'vanilla', version = VERSION, description = 'All data pulled from the website without significant modifications apart from the web page structure parsing'),
|
34 |
+
BuilderConfig(name = 'pulled', version = VERSION, description = 'Same as vanilla, but sound files have been pulled from the website, and "source" command is replaced with "sound" column')
|
35 |
]
|
36 |
|
37 |
DEFAULT_CONFIG_NAME = 'quotes'
|
|
|
61 |
'text': Value('string'),
|
62 |
'champion': Value('string')
|
63 |
})
|
64 |
+
elif self.config.name == 'pulled':
|
65 |
+
features = Features({
|
66 |
+
'header': Value('string'),
|
67 |
+
'subheader': Value('string'),
|
68 |
+
'text': Value('string'),
|
69 |
+
'sound': Audio(sampling_rate = 44_100),
|
70 |
+
'champion': Value('string')
|
71 |
+
})
|
72 |
+
else:
|
73 |
+
raise ValueError(f'Unknown config: {self.config.name}')
|
74 |
|
75 |
return DatasetInfo(
|
76 |
description=_DESCRIPTION,
|
|
|
88 |
SplitGenerator(
|
89 |
name = Split.TRAIN,
|
90 |
gen_kwargs = {
|
91 |
+
"path": dl_manager.download_and_extract(url),
|
92 |
+
'sound': dl_manager.download_and_extract(_SOUND_URL) if name == 'pulled' else None
|
93 |
}
|
94 |
)
|
95 |
]
|
96 |
|
97 |
+
def _generate_examples(self, path: str, sound: str):
|
98 |
for i, row in read_csv(path, sep = '\t').iterrows():
|
99 |
+
if sound is None:
|
100 |
+
yield i, dict(row)
|
101 |
+
else:
|
102 |
+
data = dict(row)
|
103 |
+
|
104 |
+
folder = data['folder']
|
105 |
+
filename = data['filename']
|
106 |
+
|
107 |
+
if folder == folder and filename == filename: # if folder and filename are not nan
|
108 |
+
data['sound'] = os.path.join(sound, folder, f'{filename}.ogg')
|
109 |
+
else:
|
110 |
+
data['sound'] = NA
|
111 |
+
|
112 |
+
data.pop('folder')
|
113 |
+
data.pop('filename')
|
114 |
+
|
115 |
+
yield i, data
|