fix(eval): replace a re-run model's stale samples - #45
Merged
Conversation
Dropping --n left the tail of a larger previous run in place, and a sample that errored then succeeded kept its .error.txt beside the new html, counting once as errored and once as scored. Only files the runner writes are removed, and only for models measured in this invocation.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Re-running a model did not clear its condition directories. Dropping
--nleft the tail of a larger previous run in place, and a sample that errored last time but succeeded now kept its.error.txtbeside the new.html, counting once as errored and once as scored.Carry-forward is a per-model decision, so a model being re-run is being replaced. Files matching the three names the runner writes are removed for the cells measured in this invocation, and a count is reported. Anything else in the directory survives, and models not re-run are untouched, which is what keeps one-model-at-a-time working.
Covers three behaviours in test: the tail after lowering n, the stale error sibling after a success, and preservation of both an unrelated model and a hand-placed file.