Skip to content

Conversation

@jemeza-codegen
Copy link
Contributor

Added subsets of swe-bench lite that are designed to match the distribution of the dataset.

Motivation

We need quick and reliable methods to iterate on the agent.

Content

  • adds subsets of the lite dataset designed to match the distribution of the entire dataset. Three datasets were added of lengths 10, 20 and 100.
  • added command line support
  • the agent traces are tagged with the difficulty of the task

Testing

This is a test.

Please check the following before marking your PR as ready for review

  • I have added tests for my changes
  • I have updated the documentation or added new documentation as needed

@jemeza-codegen jemeza-codegen requested review from a team and codegen-team as code owners March 8, 2025 00:08
@jemeza-codegen jemeza-codegen merged commit b763d1f into develop Mar 8, 2025
17 of 18 checks passed
@jemeza-codegen jemeza-codegen deleted the jmeza-swebench-subsets branch March 8, 2025 00:22
@github-actions
Copy link
Contributor

github-actions bot commented Mar 9, 2025

🎉 This PR is included in version 0.48.5 🎉

The release is available on GitHub release

Your semantic-release bot 📦🚀

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants