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

Breaking-IID baselines with global imbalance but w/o chunking #26

Open
mcw92 opened this issue Jan 7, 2025 · 0 comments
Open

Breaking-IID baselines with global imbalance but w/o chunking #26

mcw92 opened this issue Jan 7, 2025 · 0 comments
Assignees
Labels
experiments Tasks related to experiments to run

Comments

@mcw92
Copy link
Member

mcw92 commented Jan 7, 2025

Start runs for globally imbalanced datasets without chunking, i.e., with the globally_imbalanced flag set and mu_data = 0.5, 2.0, inf but the locally_imbalanced flag not set. These runs serve as a baseline for the breaking-IID experiments which use chunking of the global train data over all available nodes by default.

@mcw92 mcw92 added the experiments Tasks related to experiments to run label Jan 7, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
experiments Tasks related to experiments to run
Projects
None yet
Development

No branches or pull requests

2 participants