Flight Log / microsoft/promptpex #179
Fix NaN in Results Overview when eval metric score is not a number — a GitHub Copilot (GitHub) bot flight
The GitHub Copilot bot (GitHub) opened this pull request in microsoft/promptpex on Jun 20, 2025; a burn — closed unmerged after human review: +15,891 −14,989, 117 files, 2 reviews.
Flown by
- Lines
- +15,891 −14,989
- Files
- 117
- Commits
- 3
- Reviews
- 2
- To merge
- —
- Task
- fix
Line and file counts are summed per commit from the AIDev dataset; GitHub's net diff can be smaller.
GitHub Copilot attempted a fix in microsoft/promptpex for NaN appearing in the Results Overview table when an eval metric returns a non-numeric score. The change spanned 117 files, with 15891 additions and 14989 deletions, across three commits. The reviewer wrote "merge with latest 'dev' branch". The pull request closed unmerged after two reviews.
“merge with latest 'dev' branch”
— from the review (source)
From the pull request description
When eval metrics return non-numeric scores (like "❓" for unknown results), the Results Overview table incorrectly shows "NaN" instead of the appropriate fallback value. ## Problem The issue occurs in the `computeOverview` function when: 1. An eval metric returns a non-numeric response (e.g., "❓",…
Excerpt (first 300 characters) — the full description is on GitHub.
View the pull request on GitHub →
This is your flight? Claim it — sign in with GitHub. Ownership is proven by the sign-in: the account that opened the PR is the only one that can claim it.
Flight timeline
- —committedcopilot-swe-agent[bot]
- —committedcopilot-swe-agent[bot]
- —reviewed
- —reviewed
- —committedcopilot-swe-agent[bot]
- Jul 18, 2025closedpelikhan
Read next
Cite this flight: AgentRiders Flight Log, microsoft/promptpex#179, record updated Nov 3, 2026, https://agentriders.com/logs/microsoft/promptpex/179/
Data from public GitHub activity; this page is not affiliated with GitHub or the repository.