David Nicholson

Ah right of course there's no csv because you're telling me that `vak prep` crashed. Got it. And I left out some variable definitions from that snippet. This is why...

Ok, not sure what's going on. Can we try tech support through Zoom in the next couple days? Please email me. I think I need a little more context that's...

Thank you @avanikop Notes on this: - the error message should print the file names to be more informative (and anything else that could be useful). The paths are already...

Should probably also change main how-to page so it says something like "if you're using a common format like Praat TextGrid or Audacity LabelTracks, it's already built-in; follow this how-to....

yes, I agree that it might be useful to provide a combined / compound loss with hyperparameters that weight each of the loss terms. You mean something like this? https://github.com/NickleDave/vak/blob/f99ffe17ebd915b43e7f235ba5ecb127f634619e/src/vak/nn/modules/loss.py#L41...

We would need to change the name of the checkpoint saved on the validation step to something more generic, e.g. `checkpoint.best.pt`, which would be a breaking change

Goal of this is to make it easier to instantiate a model, and fix #362 (linking here so I can close that one)

The one thing this still does not give us is a `__call__` dunder-method, in the same way one can do with a `torch.Module` though :( Would be really nice to...

I am definitely not an expert in this area but isn't this a case where `cibuildwheel` could help? https://cibuildwheel.readthedocs.io/en/stable/#usage Many examples here: https://cibuildwheel.readthedocs.io/en/stable/working-examples/

I know this issue a bit stale but I want to agree with @joelb123 and echo the points made. As someone developing a library, not an application, I would love...