fix(loader): updated the format of entries in the loader script
Browse files
batch.py
CHANGED
@@ -19,7 +19,7 @@ _URLS = {
|
|
19 |
}
|
20 |
_INDEX = 'https://huggingface.co/datasets/zeio/batch/resolve/main/index.tsv'
|
21 |
|
22 |
-
_N_ITEMS =
|
23 |
_N_BATCH = 20
|
24 |
|
25 |
|
@@ -49,21 +49,21 @@ class Batch(GeneratorBasedBuilder):
|
|
49 |
if self.config.name == 'written':
|
50 |
features = Features({
|
51 |
'title': Value('string'),
|
52 |
-
'topics':
|
53 |
-
'posts':
|
54 |
'text': Value('string')
|
55 |
-
}
|
56 |
-
}
|
57 |
})
|
58 |
elif self.config.name == 'spoken':
|
59 |
features = Features({
|
60 |
'title': Value('string'),
|
61 |
'speech': Audio(sampling_rate = 48_000),
|
62 |
-
'topics':
|
63 |
-
'posts':
|
64 |
'text': Value('string')
|
65 |
-
}
|
66 |
-
}
|
67 |
})
|
68 |
else:
|
69 |
raise ValueError(f'Unknown config: {self.config.name}')
|
@@ -121,7 +121,10 @@ class Batch(GeneratorBasedBuilder):
|
|
121 |
for i, row in read_csv(index, sep = '\t').iterrows():
|
122 |
# print(row)
|
123 |
|
124 |
-
|
|
|
|
|
|
|
125 |
|
126 |
topics = []
|
127 |
posts = []
|
@@ -150,9 +153,7 @@ class Batch(GeneratorBasedBuilder):
|
|
150 |
'topics': topics
|
151 |
}
|
152 |
|
153 |
-
if spoken is None:
|
154 |
-
item['speech'] = None
|
155 |
-
else:
|
156 |
speech_cluster_path = spoken.get(row['path'])
|
157 |
|
158 |
if speech_cluster_path is None:
|
|
|
19 |
}
|
20 |
_INDEX = 'https://huggingface.co/datasets/zeio/batch/resolve/main/index.tsv'
|
21 |
|
22 |
+
_N_ITEMS = 1750
|
23 |
_N_BATCH = 20
|
24 |
|
25 |
|
|
|
49 |
if self.config.name == 'written':
|
50 |
features = Features({
|
51 |
'title': Value('string'),
|
52 |
+
'topics': [{
|
53 |
+
'posts': [{
|
54 |
'text': Value('string')
|
55 |
+
}]
|
56 |
+
}]
|
57 |
})
|
58 |
elif self.config.name == 'spoken':
|
59 |
features = Features({
|
60 |
'title': Value('string'),
|
61 |
'speech': Audio(sampling_rate = 48_000),
|
62 |
+
'topics': [{
|
63 |
+
'posts': [{
|
64 |
'text': Value('string')
|
65 |
+
}]
|
66 |
+
}]
|
67 |
})
|
68 |
else:
|
69 |
raise ValueError(f'Unknown config: {self.config.name}')
|
|
|
121 |
for i, row in read_csv(index, sep = '\t').iterrows():
|
122 |
# print(row)
|
123 |
|
124 |
+
try:
|
125 |
+
path = os.path.join(written[row['path']], f'{row["thread"]}.txt')
|
126 |
+
except KeyError:
|
127 |
+
break
|
128 |
|
129 |
topics = []
|
130 |
posts = []
|
|
|
153 |
'topics': topics
|
154 |
}
|
155 |
|
156 |
+
if spoken is not None:
|
|
|
|
|
157 |
speech_cluster_path = spoken.get(row['path'])
|
158 |
|
159 |
if speech_cluster_path is None:
|