ModelEvaluation

Intro

Often a method is needed to evaluate the quality of a policy, ModelEvaluation.py is the script to do this. This script will execute one round of model evaluation. The amount and length of the simulations the model is evaluated over depends on the values in the config file.

Example

python3 ModelEvaluation.py --config=settings/terrainRLImitate/CACLA/TL_Only_Distillation.json --sim_config_file=./args/genBiped2D/biped2d_full_gaps_with_terrain_features.txt --data_folder=Biped2dFull/TL_Only/Flat_Incline_Steps_Slopes_Gaps_3/_0/