mrq/DL-Art-School
1
3
Fork 5
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix collating on by default on grand_conjoined

Browse Source
This commit is contained in:
James Betker 2022-01-01 10:30:15 -07:00
parent 4d9ba4a48a
commit d5a5111890
2 changed files with 11 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
codes/data/audio/grand_conjoined_dataset.py
View File

@ -174,15 +174,15 @@ if __name__ == '__main__':
'batch_size': batch_sz,
'max_paired_audio_length': 255995,
'max_paired_text_length': 200,
'max_solo_text_length': 330,
'max_solo_audio_length': 300000,
'needs_collate': False,
'max_paired_text_length': 100,
'max_solo_text_length': 200,
'max_solo_audio_length': 307195,
'num_conditioning_candidates': 1,
'conditioning_length': 44000,
'needs_collate': True,
'paired_dataset_args': {
'path': ['Y:\\clips\\podcasts-0-transcribed.tsv'],
'fetcher_mode': ['tsv'],
'path': ['Z:\\bigasr_dataset\\tedlium\\train-all.txt'],
'fetcher_mode': ['libritts'],
'use_bpe_tokenizer': False,
},
'unsupervised_audio_args': {

5
codes/data/zero_pad_dict_collate.py
View File

@ -35,8 +35,11 @@ class ZeroPadDictCollate():
first_dict = batch[0]
collated = {}
for key in first_dict.keys():
if isinstance(first_dict[key], torch.Tensor) and len(first_dict[key].shape) > 0:
if isinstance(first_dict[key], torch.Tensor):
if len(first_dict[key].shape) > 0:
collated[key] = self.collate_tensors(batch, key)
else:
collated[key] = torch.stack(batch[key])
else:
collated[key] = self.collate_into_list(batch, key)
return collated