archinetai / audio-diffusion-pytorch

Audio generation using diffusion models, in PyTorch.
MIT License
1.95k stars 169 forks source link

Add trainer #3

Closed nateraw closed 2 years ago

nateraw commented 2 years ago

Hey there, I have been following this project pretty closely...looks great. Could you share the lightning trainer you're using here/any associated scripts for training?

I've put together my own trainer with accelerate which is working fine, but would be nice to work out of the same one as you here for reproducibility sake.

flavioschneider commented 2 years ago

Thanks! It's something I have in my to-do list. I was thinking about adding a separate repository with the trainer to avoid polluting this one – the one I'm currently using has a lot of extra code that I need to refactor out.

nateraw commented 2 years ago

That's sorta what I figured! A separate repo or even just a separate folder here that's not included in the python package would be great. Happy to help wherever needed.

flavioschneider commented 2 years ago

Let me know if this works for you audio-diffusion-pytorch-trainer!

nateraw commented 2 years ago

Will have a look and follow up there! Thanks 😄