Repository navigation
Measured: v1.3.0 once on Lite/Verified; function-level denominator fix - #143
Merged
Merged
Conversation
…nator fix Lite strict 276: Acc@1/3/5/10 0.507/0.717/0.750/0.815 (v1.2.0 0.486/0.710/0.750/0.804). LocAgent 274: 0.500/0.723/0.755; function Acc@5/@10 0.394/0.482 (BM25 0.318/0.369). Verified 500: 0.446/0.680/0.736/0.814; not in Lite (403): 0.437/0.670/0.732/0.811. func_eval.py counted only instances that have a `definitions` list; short reports (<60 words) get none, so they silently left the denominator (244 instead of 274, Acc@5 0.443 instead of 0.394). They now count as misses; --skip-missing keeps the old rule. Adds --ci and --gold-cache; git show failures retry and raise. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Phase 0 of session 18 (docs + scoring script only, no engine change — the 14 gold sets measure the unchanged binary).
v1.3.0 release binary, run once (95% bootstrap CI):
Function level, LocAgent 274: Acc@1/5/10 0.168/0.394/0.482 (BM25† 0.318/0.369 @5/@10, CodeRankEmbed† 0.518/0.588).
Scoring fix:
func_eval.pydropped instances without adefinitionslist (reports < 60 words) from the denominator — 244 instead of 274, Acc@5 0.443 instead of 0.394. Now they count as misses (--skip-missingkeeps the old rule). The 274 now equals LocAgent's subset exactly.Phase 2 (strong-LLM stage) is blocked: all five supplied Baseten keys return HTTP 402 (no credit).
🤖 Generated with Claude Code