Unclear error message thrown by stft if input and window arguments are on different devices about gard-adversarial-speaker-id HOT 1 CLOSED

Traceback (most recent call last):
File "train_libri.py", line 238, in
main(parse_args())
File "train_libri.py", line 113, in main
real_feature = model.encode(inputs)
File "/home/ics/gard-adversarial-speaker-id-master/dev/models/model.py", line 84, in encode
x = self.prep(x.squeeze(1))
File "/home/ics/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "/home/ics/gard-adversarial-speaker-id-master/dev/transforms/dataset.py", line 111, in forward
window=torch.hann_window(hp.win_length))
File "/home/ics/anaconda3/envs/pytorch/lib/python3.7/site-packages/torch/functional.py", line 472, in stft
normalized, onesided, return_complex)
RuntimeError: stft input and window must be on the same device but got self on cuda:0 and window on cpu

from gard-adversarial-speaker-id.