Skip to content

Conversation

@barvhaim
Copy link
Contributor

@CLAassistant
Copy link

CLAassistant commented Aug 21, 2024

CLA assistant check
All committers have signed the CLA.

Copy link
Contributor

@greptile-apps greptile-apps bot left a comment

Choose a reason for hiding this comment

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

Disclaimer: Experimental PR review

PR Summary

This pull request adds support for the Watsonx.ai model in the Langfuse Python SDK, specifically for the Langchain extension. The changes span multiple files to integrate Watsonx.ai functionality into the existing codebase.

Key changes include:

  • Added Watsonx.ai model parameter handling in langfuse/callback/langchain.py
  • Extended model name extraction for Watsonx.ai in langfuse/extract_model.py
  • Modified usage metric conversion in langfuse/utils/__init__.py to support Watsonx.ai format
  • Updated sampling logic in langfuse/Sampler.py to handle Watsonx.ai event structures

Important considerations:

  • Ensure comprehensive testing for Watsonx.ai integration, including edge cases
  • Verify that changes in Sampler.py don't affect existing functionality for other models
  • Consider adding explicit type hints and docstring updates for new Watsonx.ai-related code
  • Evaluate if any additional error handling or logging is needed for Watsonx.ai-specific scenarios

4 file(s) reviewed, no comment(s)
Edit PR Review Bot Settings

@maxdeichmann maxdeichmann changed the base branch from main to watsonx-llm August 28, 2024 15:01
@maxdeichmann maxdeichmann merged commit d294918 into langfuse:watsonx-llm Aug 28, 2024
maxdeichmann added a commit that referenced this pull request Aug 28, 2024
Support for Watsonx ai model for langchain extension (#874)

Co-authored-by: Bar Haim <barvhaim@gmail.com>
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.

3 participants