-
Notifications
You must be signed in to change notification settings - Fork 5.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat: add syncing models utility to ivy #28818
base: main
Are you sure you want to change the base?
Conversation
My thinking was just exposing these as |
@YushaArif99 Couldn't we just store all helper functions like this in a new file like Also, I'd suggest we could expose a general |
Sure! This is indeed a better UX. I was only inclined to not go this route as the logic contained within these utility helpers is framework-specific. And so it seems to deviate from the general convention of Otherwise, this is definitely cleaner. Based on this, should we still go forward with having these inside |
Yeah that's a good idea 👍🏼 |
@YushaArif99 I don't see an issue with it, because you'd have to import torch into the tensorflow backend anyway, which is also against the general convention of ivy. But I guess if we do expose |
Yeah I think exposing I'll go ahead with this approach then. Thanks for the suggestions both! |
Wanted to have your thoughts on how to expose these utility functions? I think it makes sense to have them inside the
ivy.functional.backends.tensorflow.module.py
module. But this creates an issue as to how we import this helper. So there wont be anyivy.sync_models_torch_and_tf
but instead we'll have to do:What do you guys suggest we do here @Sam-Armstrong @hmahmood24