truera / trulens

Evaluation and Tracking for LLM Experiments
https://www.trulens.org/
MIT License
2.17k stars 189 forks source link

human feedback with metadata #1629

Closed sfc-gh-pmardziel closed 1 week ago

sfc-gh-pmardziel commented 2 weeks ago

Description

Expanded human_feedback.ipynb example to also show how to add metadata to the feedback. Also improved the original example in various ways.

Adjusted FeedbackCall schema to include some default values to make it easier to specify metadata only.

Other details good to know for developers

Please include any other details of this change useful for TruLens developers.

Type of change


[!IMPORTANT] Expanded human_feedback.ipynb to include metadata and improved feedback handling in sqlalchemy.py and feedback.py.

  • Behavior:
    • Expanded human_feedback.ipynb to include metadata in feedback.
    • Improved the original example with unspecified enhancements.
  • Code Changes:
    • In sqlalchemy.py, changed feedback name construction to use core_db.MULTI_CALL_NAME_DELIMITER.
    • In feedback.py, set default values for args and ret in FeedbackCall using pydantic.Field.

This description was created by Ellipsis for b120c627d65a154b98345aba4801eb3a05bb0fdc. It will automatically update as commits are pushed.

review-notebook-app[bot] commented 2 weeks ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.


Powered by ReviewNB