Fix scoring false positives and infrastructure reliability#40
Closed
abrichr wants to merge 1 commit into
Closed
Conversation
- Guard empty metric_results in evaluate_endpoint.py to prevent min([])/all([]) crashes and false positives - Add error_type field to BenchmarkResult for distinguishing infrastructure vs agent vs evaluation failures - Set error_type="infrastructure" on evaluation timeout and request errors - Use config.timeout instead of hardcoded 30s for screenshot/a11y requests - Remove reasonable_completion false-positive path in mock adapter (calling done after 2+ actions no longer counts as success) - Health check now probes both WAA server and evaluate server Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Member
Author
|
Superseded by #42 (combined PR targeting main) |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
metric_resultsin evaluate_endpoint.py to preventmin([])/all([])crashes and false positiveserror_typefield toBenchmarkResultfor distinguishing infrastructure vs agent vs evaluation failureserror_type="infrastructure"on evaluation timeout and request errors in live adapterconfig.timeoutinstead of hardcoded 30s for screenshot/a11y requestsreasonable_completionfalse-positive path in mock adapter (calling done after 2+ actions no longer counts as success)run_dc_eval.pynow probes both WAA server (port 5001) and evaluate server (port 5050)Test plan
openadapt-evals mock --tasks 5to confirm mock no longer gives false successeserror_typeis populated🤖 Generated with Claude Code