EOFError: Ran out of input running any example

[NoiseGener8r]

I have tried deleting the .cache folder and its contents several times, as well as running all examples and trying to run it with my own music.

Traceback (most recent call last): File "jukebox/sample.py", line 279, in <module> fire.Fire(run) File "C:\Users\rfnoi\.conda\envs\jukebox2\lib\site-packages\fire\core.py", line 127, in Fire component_trace = _Fire(component, args, context, name) File "C:\Users\rfnoi\.conda\envs\jukebox2\lib\site-packages\fire\core.py", line 366, in _Fire component, remaining_args) File "C:\Users\rfnoi\.conda\envs\jukebox2\lib\site-packages\fire\core.py", line 542, in _CallCallable result = fn(*varargs, **kwargs) File "jukebox/sample.py", line 276, in run save_samples(model, device, hps, sample_hps) File "jukebox/sample.py", line 181, in save_samples vqvae, priors = make_model(model, device, hps) File "c:\users\rfnoi\jukebox\jukebox\make_models.py", line 191, in make_model vqvae = make_vqvae(setup_hparams(vqvae, dict(sample_length=hps.get('sample_length', 0), sample_length_in_seconds=hps.get('sample_length_in_seconds', 0))), device) File "c:\users\rfnoi\jukebox\jukebox\make_models.py", line 95, in make_vqvae restore_model(hps, vqvae, hps.restore_vqvae) File "c:\users\rfnoi\jukebox\jukebox\make_models.py", line 55, in restore_model checkpoint = load_checkpoint(checkpoint_path) File "c:\users\rfnoi\jukebox\jukebox\make_models.py", line 37, in load_checkpoint checkpoint = t.load(restore, map_location=t.device('cpu')) File "C:\Users\rfnoi\.conda\envs\jukebox2\lib\site-packages\torch\serialization.py", line 387, in load return _load(f, map_location, pickle_module, **pickle_load_args) File "C:\Users\rfnoi\.conda\envs\jukebox2\lib\site-packages\torch\serialization.py", line 564, in _load magic_number = pickle_module.load(f, **pickle_load_args) EOFError: Ran out of input

[zdhoward]

I keep hitting this issue on Windows too. Deleting the .cache doesn't seem to change anything.

I experience this anytime I run any example or I try to sample my own audio files.

[nbinobied]

I had similar issue with the start of training the vqvae, I was able to fix it by changing the number of workers in train.py in line number 300 from hps.bs_sample = hps.nworkers = hps.bs to hps.bs_sample = hps.nworkers = 0