Skip to content

Conversation

@juanmichelini
Copy link
Collaborator

Summary

When running eval_infer in swebenchmultimodal with no valid entries to convert, the script was crashing with a ValueError. This PR removes that exception so the script continues normally, letting the harness be responsible for handling empty results.

Changes

  • Removed the ValueError exception in convert_to_swebench_format when converted_count == 0
  • Added a test to verify empty input files don't raise exceptions

Fixes #338

@juanmichelini can click here to continue refining the PR

When convert_to_swebench_format converts 0 entries, it should not raise
an exception. Instead, let the harness be responsible for handling the
empty results.

Fixes #338
@openhands-ai openhands-ai bot mentioned this pull request Jan 18, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

SWE-Bench Multimodal

3 participants