Skip to content

Conversation

@lxdlam
Copy link
Contributor

@lxdlam lxdlam commented Mar 24, 2025

The original ProgramOfThought module was excellent but limited to a single output field. This PR refines the module to support multiple return values while incorporating improvements from PR #7992.

Changes include:

Backward compatibility:

  • The original one-line PoT answer style remains supported.

Tests:

  • Added tests to cover these changes.
  • Verified functionality with local models (qwen2.5-14b and deepseek-r1-14b).

@lxdlam
Copy link
Contributor Author

lxdlam commented Mar 24, 2025

Hi @okhat and @TomeHirata, please review it if you have time.

)
# Currently, the interpreter class checks the deno availability at execution time.
# We may consider checking it at the initialization time for better instruction.
# It will raises exception when dspy cannot find available deno instance by now.
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This comment will be fulfilled by a new PR later.

@okhat
Copy link
Collaborator

okhat commented Mar 24, 2025

Thanks so much @lxdlam !

@okhat okhat merged commit 89770d3 into stanfordnlp:main Mar 24, 2025
4 checks passed
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.

2 participants