Error in export when using output_mode='raw'

[Bloos]

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