Skip to content

Conversation

@skyw
Copy link
Contributor

@skyw skyw commented Oct 17, 2025

No description provided.

@copy-pr-bot
Copy link

copy-pr-bot bot commented Oct 17, 2025

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@skyw skyw force-pushed the skyw/kl_shampoo_dev branch from b9357f8 to 25200cc Compare October 17, 2025 21:55
@skyw
Copy link
Contributor Author

skyw commented Oct 17, 2025

@mkhona-nvidia we need to think about how to test this.

Copy link
Contributor

@mkhona-nvidia mkhona-nvidia left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggestion: instead in the init make a kronecker_factor_update_fn and assign either update_kronecker_factors or update_kronecker_factors_kl_shampoo

need to think about how to deal with that extra argument eigenbasis_list

@skyw skyw force-pushed the skyw/kl_shampoo_dev branch from 54f7720 to 4123d53 Compare October 21, 2025 17:00
@skyw skyw marked this pull request as ready for review October 21, 2025 17:00
mkhona-nvidia
mkhona-nvidia previously approved these changes Oct 21, 2025
Copy link
Contributor

@mkhona-nvidia mkhona-nvidia left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@skyw
Copy link
Contributor Author

skyw commented Oct 21, 2025

/ok to test 4123d53

@skyw
Copy link
Contributor Author

skyw commented Oct 21, 2025

/ok to test f3509f4

@skyw skyw merged commit bc13a6b into main Oct 21, 2025
14 checks passed
@skyw skyw deleted the skyw/kl_shampoo_dev branch October 21, 2025 21:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants