Will Berman

Hey @zsyOAOA, super cool model! It might be helpful to open an issue first where we discuss the model components and what code changes we need to add them. Similar...

awesome, thanks for the context @zsyOAOA ! for auxiliary models, we have a few options 1. the auxiliary model can be used to pre-process input such that the processed input...

@isamu-isozaki I'm not sure I see what logic would have to be moved to a function? Regardless, I do think importing wandb on the top level of the module is...

Looks basically there :)

I think @dasayan05 is correct here. My understanding is that the underlying parameters in the `prepared` model and the `unwrapped` model are still the same. EMAModel.copy_to will then copy the...

I think this is still a bug as `store()` and `restore()` are not called in train_unconditional

@rmac85 thanks for the issue! I might need some more specifics on what you're asking. I think one of the schedulers you're talking about is the scheduler in the trained...

Yes, this is to be expected. One of the schedulers is used to handle adding noise while training, the other is used for inference. The config that actually determines the...

I'm a bit confused, I believe the "noise schedule" is the same regardless of the Scheduler class. I don't think we get anything out of adding this method to other...

Yeah these are neat ideas! I think these would be good options for community PRs, happy to open up to the community :)