You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think I've found another bug in the export module.
When using output_mode = "raw", the export of the trained model will fail with the following stack trace
Traceback (most recent call last):
File "/home/<user>/.local/bin/rave", line 8, in <module>
sys.exit(main())
File "/home/<user>/.local/lib/python3.10/site-packages/scripts/main_cli.py", line 38, in main
app.run(export.main)
File "/home/<user>/.local/lib/python3.10/site-packages/absl/app.py", line 308, in run
_run_main(main, args)
File "/home/<user>/.local/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main
sys.exit(main(argv))
File "/home/<user>/.local/lib/python3.10/site-packages/scripts/export.py", line 566, in main
scripted_rave = script_class(
File "/home/<user>/.local/lib/python3.10/site-packages/scripts/export.py", line 168, in __init__
self.register_method(
File "/home/<user>/.local/lib/python3.10/site-packages/nn_tilde/__init__.py", line 78, in register_method
y = getattr(self, method_name)(x)
File "/home/<user>/.local/lib/python3.10/site-packages/scripts/export.py", line 285, in decode
y = self.pqmf.inverse(y)
File "/home/<user>/.local/lib/python3.10/site-packages/rave/pqmf.py", line 289, in inverse
x = self.inverse_conv(x) * m
File "/home/<user>/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/home/<user>/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1527, in _call_impl
return forward_call(*args, **kwargs)
File "/home/<user>/.local/lib/python3.10/site-packages/cached_conv/convs.py", line 221, in forward
return nn.functional.conv1d(
RuntimeError: Given groups=1, weight of size [16, 16, 33], expected input[1, 1, 8224] to have 16 channels, but got 1 channels instead
which is strange, because pqmf isn't even supposed to be used. I suggest changing line 283 in the export script if self.pqmf is not None:
to if self.output_mode == "pqmf":
I didn't create a pull request because I don't know about the implications for other use cases. It fixed the error in my use case though
The text was updated successfully, but these errors were encountered:
I think I've found another bug in the export module.
When using output_mode = "raw", the export of the trained model will fail with the following stack trace
which is strange, because pqmf isn't even supposed to be used. I suggest changing line 283 in the export script
if self.pqmf is not None:
to
if self.output_mode == "pqmf":
I didn't create a pull request because I don't know about the implications for other use cases. It fixed the error in my use case though
The text was updated successfully, but these errors were encountered: