jukebox icon indicating copy to clipboard operation
jukebox copied to clipboard

NameError: name '_get_default_group' is not defined

Open chengkeng opened this issue 3 years ago • 3 comments

Hi! This happens when I want to try to train the model:

python jukebox/train.py --hps=small_vqvae --name=small_vqvae --sample_length=262144 --bs=4 --audio_files_dir=sanmp3 --labels=False --train --aug_shift --aug_blend Using cuda True Found 79 files. Getting durations self.sr=22050, min: 12, max: inf Keeping 77 of 79 files Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate {'l2': 0.016028581976191468, 'l1': 0.09432193525560592, 'spec': 2703.2033268855166} Creating Data Loader Train 486 samples. Test 55 samples Train sampler: <torch.utils.data.sampler.BatchSampler object at 0x000001C341B35FC8> Train loader: 30 Loading vqvae in train mode Parameters VQVAE:586050 Traceback (most recent call last): File "jukebox/train.py", line 345, in fire.Fire(run) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 127, in Fire component_trace = _Fire(component, args, context, name) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 366, in _Fire component, remaining_args) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 542, in _CallCallable result = fn(*varargs, **kwargs) File "jukebox/train.py", line 318, in run distributed_model = get_ddp(model, hps) File "jukebox/train.py", line 45, in get_ddp ddp = DistributedDataParallel(model, device_ids=[local_rank], output_device=local_rank, broadcast_buffers=False, bucket_cap_mb=hps.bucket) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\torch\nn\parallel\distributed.py", line 273, in init self.process_group = _get_default_group() NameError: name '_get_default_group' is not defined

My operating system is win10, is it caused by the problem of dependency installation? Thank you!

chengkeng avatar Mar 20 '21 04:03 chengkeng

I'm also having this issue... Does anyone have an idea of what may be causing this issue?

nuJack avatar Nov 14 '21 21:11 nuJack

+1 here, on windows 10.

sunzeyucmu avatar Feb 18 '22 06:02 sunzeyucmu

Hi! This happens when I want to try to train the model:

python jukebox/train.py --hps=small_vqvae --name=small_vqvae --sample_length=262144 --bs=4 --audio_files_dir=sanmp3 --labels=False --train --aug_shift --aug_blend Using cuda True Found 79 files. Getting durations self.sr=22050, min: 12, max: inf Keeping 77 of 79 files Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate Estimating duration from bitrate, this may be inaccurate {'l2': 0.016028581976191468, 'l1': 0.09432193525560592, 'spec': 2703.2033268855166} Creating Data Loader Train 486 samples. Test 55 samples Train sampler: <torch.utils.data.sampler.BatchSampler object at 0x000001C341B35FC8> Train loader: 30 Loading vqvae in train mode Parameters VQVAE:586050 Traceback (most recent call last): File "jukebox/train.py", line 345, in fire.Fire(run) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 127, in Fire component_trace = _Fire(component, args, context, name) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 366, in _Fire component, remaining_args) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\fire\core.py", line 542, in _CallCallable result = fn(*varargs, **kwargs) File "jukebox/train.py", line 318, in run distributed_model = get_ddp(model, hps) File "jukebox/train.py", line 45, in get_ddp ddp = DistributedDataParallel(model, device_ids=[local_rank], output_device=local_rank, broadcast_buffers=False, bucket_cap_mb=hps.bucket) File "C:\Users\Creator\miniconda3\envs\jukebox\lib\site-packages\torch\nn\parallel\distributed.py", line 273, in init self.process_group = _get_default_group() NameError: name '_get_default_group' is not defined

My operating system is win10, is it caused by the problem of dependency installation? Thank you!

Your os needs to be win8 or win8.1: https://www.microsoft.com/en-us/software-download/windows8ISO

cicinwad avatar Mar 31 '22 17:03 cicinwad