[awais-ur-rehman]

What this changes

Classification failures now list the allowed labels in suite order, using the existing normalization setting and showing at most ten entries. The result details and pass/fail rules stay the same.

Closes #40. Related: #35, which was closed so this issue could be assigned to a first-time contributor.

How it was tested

• python -m pytest: 181 passed.
• ruff check src tests examples: All checks passed!
• ruff format --check src tests examples: 21 files already formatted.
• Five regression cases fail on the original implementation and pass with the change, covering order, normalization on/off, the ten-label boundary and truncation.
• git diff --check: clean.

Checklist

• pytest passes locally
• ruff check and ruff format --check pass
• New behavior has outcome tests and a failing-before regression
• README updated for the failure message
• Tests run offline with no model downloads

OpenAI Codex assisted with implementation and test execution.

[github-actions]

Thanks for your first pull request to ModelDiff! A maintainer will review it soon. CI for first-time contributors starts after a maintainer approves it.

While you wait, please check that:

• The issue was assigned to you before you started (see One issue at a time).
• The description links the issue (Closes #12) and says how you tested the change.
• pytest, ruff check src tests examples and ruff format --check src tests examples pass locally.
• The Files changed tab shows only changes for this issue.

[AdityaWagh06]

Merged, thank you @awais-ur-rehman, and congrats on your first contribution to ModelDiff! 🎉 The message matches the issue exactly, and the tests cover the cap, normalization and the unchanged details. Clean work. It’ll ship in v0.2.0 next week.

If you enjoyed contributing, a ⭐ on the repo helps others find the project. Follow OpenWave India on GitHub and LinkedIn for new projects coming soon.