Skip to content
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

NNX WeightNorm #4426

Open
rbflx opened this issue Dec 10, 2024 · 2 comments
Open

NNX WeightNorm #4426

rbflx opened this issue Dec 10, 2024 · 2 comments
Labels
Priority: P2 - no schedule Best effort response and resolution. We have no plan to work on this at the moment. Status: pull requests welcome We agree with the direction proposed, feel free to give it a shot and file a pull request

Comments

@rbflx
Copy link

rbflx commented Dec 10, 2024

Is there an NNX implementation of the WeightNorm module as used in Linen? I can't seem to find it in the normalization layers implemented in NNX.

I'm in the process of writing a new model in NNX which would require them.

@cgarciae cgarciae added Priority: P2 - no schedule Best effort response and resolution. We have no plan to work on this at the moment. Status: pull requests welcome We agree with the direction proposed, feel free to give it a shot and file a pull request labels Dec 10, 2024
@cgarciae
Copy link
Collaborator

We don't have a WeightNorm implementation yet in NNX. It should be very easy to implement using nnx.state + nnx.statelib.map + nnx.update. Contributions welcomed.

@prakashkagitha
Copy link

Is it okay if I take a shot at this implementation? I see that NNX doesn't have InstanceNorm and SpectralNorm from Linen as well.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Priority: P2 - no schedule Best effort response and resolution. We have no plan to work on this at the moment. Status: pull requests welcome We agree with the direction proposed, feel free to give it a shot and file a pull request
Projects
None yet
Development

No branches or pull requests

3 participants