Cannot generate audio: AttributeError: 'ConvTranspose1d' object has no attribute 'padding_mode'

[vitaly-zdanevich]

$ python3 inference.py -f <(ls mel_spectrograms/*.pt) -w waveglow_256channels.pt -o . --is_fp16 -s 0.6
/opt/anaconda3/lib/python3.7/site-packages/torch/serialization.py:454: SourceChangeWarning: source code of class 'torch.nn.modules.conv.ConvTranspose1d' has changed. you can retrieve the original source code by accessing the object's source attri
bute or set `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.
  warnings.warn(msg, SourceChangeWarning)
/opt/anaconda3/lib/python3.7/site-packages/torch/serialization.py:454: SourceChangeWarning: source code of class 'torch.nn.modules.conv.Conv1d' has changed. you can retrieve the original source code by accessing the object's source attribute or s
et `torch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.
  warnings.warn(msg, SourceChangeWarning)
/opt/anaconda3/lib/python3.7/site-packages/torch/serialization.py:454: SourceChangeWarning: source code of class 'glow.Invertible1x1Conv' has changed. you can retrieve the original source code by accessing the object's source attribute or set `to
rch.nn.Module.dump_patches = True` and use the patch tool to revert the changes.
  warnings.warn(msg, SourceChangeWarning)
Selected optimization level O3:  Pure FP16 training.
Defaults for this optimization level are:
enabled                : True
opt_level              : O3
cast_model_type        : torch.float16
patch_torch_functions  : False
keep_batchnorm_fp32    : False
master_weights         : False
loss_scale             : 1.0
Processing user overrides (additional kwargs that are not None)...
After processing overrides, optimization options are:
enabled                : True
opt_level              : O3
cast_model_type        : torch.float16
patch_torch_functions  : False
keep_batchnorm_fp32    : False
master_weights         : False
loss_scale             : 1.0
Traceback (most recent call last):
  File "inference.py", line 84, in <module>
    args.sampling_rate, args.is_fp16, args.denoiser_strength)
  File "inference.py", line 54, in main
    audio = waveglow.infer(mel, sigma=sigma)
  File "/home/vitaly_zdanevich/waveglow/glow.py", line 252, in infer
    spect = self.upsample(spect)
  File "/opt/anaconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 493, in __call__
    result = self.forward(*input, **kwargs)
  File "/opt/anaconda3/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 641, in forward
    if self.padding_mode != 'zeros':
  File "/opt/anaconda3/lib/python3.7/site-packages/torch/nn/modules/module.py", line 539, in __getattr__
    type(self).__name__, name))
AttributeError: 'ConvTranspose1d' object has no attribute 'padding_mode'

My PyTorch is not 1.0 but 1.1 - in Google Cloud this version is preinstalled into the machine learning image.

Related issue: NVIDIA/tacotron2#182

[rafaelvalle]

Pytorch added padding_mode as an attribute of convs.
Try this:

for m in waveglow.modules():
    if 'Conv' in str(type(m)):
        setattr(m, 'padding_mode', 'zeros')