clp-research / clembench

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

plotting the generated graphs next to the target graph #121

Closed yanweiser closed 3 months ago

yanweiser commented 3 months ago

This extension will create a plot of the generated graph next to the target graph for each turn of each episode in the mm_mapworld_graphs game during scoring. All steps are separately stored as pdf files. Additionally a single gif image is created that shows the progress over the course of the episode.