fix(csa): preserve tracing without mutation families - #115
Merged
Merged
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Change
Generation tracing now skips mutation-weight calculation when no mutation family is configured. Previously, valid regular-only, initial-only, and mixed regular/initial schedules failed at the first generated batch with
ValueError("family must not be empty").The weight helper still rejects an empty distribution, and the trace projection still includes recorded family statistics. Sampling, adaptation, and the trace-disabled execution path are unchanged.
Regression Coverage
Before the fix, 30 new generation/checkpoint cases reproduce the failure, while eight nonempty-mutation controls pass. All pass with the fix.
The tests compare proposals, including IDs and order, RNG snapshots, and other execution state after each ask/tell over four generations. They cover adaptation and shuffling on/off, batch sizes 1 and 8, checkpoint continuation, and the distinct contracts of strict weight queries and empty sampling.