Repository navigation
Report the scaling factor when the RCP check fails - #487
Open
matthew-frank wants to merge 1 commit into
Open
matthew-frank wants to merge 1 commit into
matthew-frank wants to merge 1 commit into
Conversation
A failed RCP check writes no scaling.json, so the factor the score would have been normalized by was never shown. Log it as a warning in the fail branch of _eval_submission_record, so it appears wherever the check runs: the rcp_checker CLI, the package checker (including results that pass via rcp-bypass), the result summarizer, and compute_score. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅ |
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When a result fails the RCP check (converges faster than the RCP min), no scaling.json is written, so there is no record of the factor the score would have been normalized by. That makes it hard to review results that fail the check, especially ones that go through with
rcp-bypass.This change logs that factor as a warning in the fail branch of
_eval_submission_record, for example:Because it is in the RCP checker itself, the message appears wherever the check runs: the rcp_checker CLI, the package checker (including results that pass via
rcp-bypass), the result summarizer, and compute_score. It is logged at warning level because the result summarizer and compute_score do not configure logging, so info messages would not be shown there.No files are written and no pass/fail behavior changes. Failures where no RCP comparison happens (e.g. missing RCP above the largest batch size) are unaffected, since no factor exists for them.
🤖 Generated with Claude Code