clp-research / clembench

A Framework for the Systematic Evaluation of Chat-Optimized Language Models as Conversational Agents and an Extensible Benchmark
MIT License
25 stars 34 forks source link

changing mm_mapworld games to show images in the transcripts #108

Closed yanweiser closed 4 months ago

yanweiser commented 4 months ago

changed logging in all mm_mapworld games to show the images of the room that is being visited each turn. I had also added the missing else clause to clemgame/transcript_utils.py but I see that has already been fixed today.

sherzod-hakimov commented 4 months ago

@yanweiser : the change version of clemgame/transcript_utils.py is missing in your commit.