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

Implementing SCAFFOLD #20

Open
Saipraneet opened this issue Jan 23, 2021 · 2 comments
Open

Implementing SCAFFOLD #20

Saipraneet opened this issue Jan 23, 2021 · 2 comments
Labels
algorithm Federated algorithms contributions welcome Issues for which community contributions are welcome

Comments

@Saipraneet
Copy link

It might be a good idea to have an implementation of SCAFFOLD as well in the algorithms. I think this can be done by modifying the existing Mime implementation.

@stheertha
Copy link
Collaborator

This is a good suggestion, will keep in the radar. We are also planning to accept community contributions. Please let us know if you are interested in implementing it.

@jaehunro jaehunro added algorithm Federated algorithms contributions welcome Issues for which community contributions are welcome labels Jun 29, 2021
@jaehunro
Copy link
Collaborator

Thanks for your patience.

We are now ready to accept community contributions. We recommend reading through the contributing guidelines as well as the developer workflow.

If you're interested in implementing, add SCAFFOLD to fedjax.algorithms. Please refer to fed_avg.py and fed_avg_test.py for an example internal implementation and unit tests for FedAvg. Additionally, as mentioned in the contributing guidelines, please also add an associated binary that runs SCAFFOLD on EMNIST that reproduces the paper results.

Please reach out if you have any additional questions.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
algorithm Federated algorithms contributions welcome Issues for which community contributions are welcome
Projects
None yet
Development

No branches or pull requests

3 participants