Repository navigation
fix(sweeps): filter non-finite values so NaN/inf metrics don't crash gp.fit - #144
Merged
Merged
Conversation
This comment has been minimized.
This comment has been minimized.
kmikowicz-wandb
approved these changes
Sep 24, 2026
kmikowicz-wandb
left a comment
Contributor
There was a problem hiding this comment.
In github.com/wandb/sweeps-bench we have benchmarks on convergence. I would run about 20 runs on all of the problems there to quickly check the bayes optimization still converges on a solution to the math problems as fast as the previous version
Contributor
Author
|
ran sweeps-bench on 10 of the problems over 20 trials using fixed seeds and no differences in performance |
nathancy-wandb
enabled auto-merge (squash)
September 25, 2026 19:39
nathancy-wandb
added a commit
that referenced
this pull request
Sep 25, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fixes https://coreweave.atlassian.net/browse/WB-9541 and https://coreweave.atlassian.net/browse/WB-40181
For https://coreweave.atlassian.net/browse/WB-40181, NaN metrics themselves are imputed to the worst finite metric seen in the sweep. However if that value is large enough, the sum operation in np.mean will overflow and result in a NaN anyway. We instead scale the metrics by the largest value, compute the mean on the scaled values, then rescale.
For https://coreweave.atlassian.net/browse/WB-9541, it's possible to have NaN current_x values if user submitted sweep runs contain invalid values (such as a negative value for a uniform-log distributed hyperparameter).