Closed bgenchel closed 5 months ago
@bgenchel seems that the model checkpoint on zenodo is down
import torch
separator = torch.hub.load('sigsep/open-unmix-pytorch', 'umxhq')
produces the same error. I will investigate
fixed in newest release v1.3.0
🐛 Bug
Not a bug, but the library just doesn't seem to be able to be used outside the CLI out of the box. I'm trying to import a separator as specified in the documentation, but if you import the class through models, then you need to specify a Model (nn.Model), which you don't have access to out of the box. Using one of the util methods, load_separator, also does not work, as it leads to a 403 Forbidden Error.
To Reproduce
Steps to reproduce the behavior:
Expected behavior
I can load a separator in code and use it to create stems programmatically without calling from the CLI or using subprocess.
Environment
Please add some information about your environment
If unsure you can paste the output from the pytorch environment collection script (or fill out the checklist below manually).
You can get that script and run it with:
Additional context