Dependencies: TensorFlow, MuJoCo version 1.31, OpenAI Gym
The only file that you need to look at is run_expert.py, which is code to load up an expert policy, run a specified number of roll-outs, and save out data.
In experts/, the provided expert policies are:
- Ant-v1.pkl
- HalfCheetah-v1.pkl
- Hopper-v1.pkl
- Humanoid-v1.pkl
- Reacher-v1.pkl
- Walker2d-v1.pkl
The name of the pickle file corresponds to the name of the gym environment.
The code depends on TensorFlow r12.0.
Simply run
bash run_all.shSeveral new folders will be generated and you can find all the figures under figures directory.