Fabian-Robert Stöter comments

Results 398 comments of


                                            Fabian-Robert Stöter

Model download is very slow

I acknowledge it's getting slower and slower. Will consider using huggingfaces for at least the umxl model

Model download is very slow

i think download speed will be improved in the new release

Hub loader doesn't use wiener_win_len argument value

@papahabla good catch. Are you able to provide a PR to fix this?

RuntimeError: "reflection_pad1d_out_template" not implemented for 'Short' : when using separate(...) method

@deepakpawade the current master version doesn't support 1.10 yet. The tests still run on torch 1.9. See #112

RuntimeError: "reflection_pad1d_out_template" not implemented for 'Short' : when using separate(...) method

@deepakpawade can this be closed?

Train single channel model using left or right channel

@aliutkus you agree?

feat: differentiable PyTorch backend

@yoyololicon thats great 🙌 Have you seen our implementation in https://github.com/sigsep/open-unmix-pytorch/blob/master/openunmix/filtering.py ? - Is yours using complex or real dtypes? - can you benchmark against the one in openunmix? -...

feat: differentiable PyTorch backend

@yoyololicon sorry for the slow response. I will have a look next week 🙏

feat: differentiable PyTorch backend

@yoyololicon Totally forgot about this one. I will have a look again

IndexError when evaluate by museval

Can you check the duration and number of channels? Maybe your predictions are mono instead of stereo?