Skip to content

Releases: trubrics/trubrics-python

Release 1.8.5

31 Mar 07:36
Compare
Choose a tag to compare

Added

  • prompt_event_name and generation_event_name to track_llm_events

Release 1.8.4

17 Mar 08:11
Compare
Choose a tag to compare

Added

  • thread_id to track_llm_events

Changed

  • assistant_id now optional in track_llm_events

Release 1.8.3

26 Feb 11:10
4b73474
Compare
Choose a tag to compare

Changed

  • Change logging level to debug for flushing events and adding event to queue

Added

  • LLM events are now routed to the new publis_llm_events endpoint

Release 1.8.2

11 Feb 14:05
Compare
Choose a tag to compare

Fixed

  • Added missing flushing limit values, and setup automated flush if limit is reached

Changed

  • Change LLM tracking latency to seconds and remove default latency from event properties

Release 1.8.1

08 Feb 07:29
6344a4e
Compare
Choose a tag to compare

Fixed

  • Fixed error logging when flushing events

Release 1.8.0

06 Feb 14:37
01c2f3d
Compare
Choose a tag to compare

Added

  • Update logging, removing is_verbose parameter
  • Removed CONTRIBUTING.md, as info is now in README.md
  • Brought back repository to PyPI, and uses uv to build and publish
  • Changed name to "trubrics" for easier installation

v1.6.2

24 Oct 09:21
Compare
Choose a tag to compare

Added

  • Option to user text_area for textual feedback collection in Streamlit. Thanks @hamdan-27

v1.6.1

12 Sep 10:19
Compare
Choose a tag to compare

Fixed

  • Now compatible for pydantic>=1.5 (v1 and v2)

v1.6.0

08 Sep 08:17
Compare
Choose a tag to compare

streamlit-feedback has been revamped for tighter integration with st chat elements

Changed

  • fix dependency with streamlit-feedback==0.1.2
  • single_submit has been removed from .st_feedback(), and replaced with disable_with_score

Added

  • new streamlit chatbot streaming example

Fixed

  • streamlit llm examples

v1.5.2

04 Sep 14:32
Compare
Choose a tag to compare

Fixed

  • user_response is dict rather than Response for feedback_type="text"
  • collector.st_feedback(..., save_to_trubrics=False) returns a python dict
  • pydantic dependency updated and .dict() updated to .model_dump()