Reinforcement Learning Summer School : Practical Tutorial on RL for Continuous Control
Here we go over:
- How to setup MuJoCo and openai/rllab
- How to run basic TRPO and DDPG code
- The core code snippets in TRPO and DDPG so you can build on top of these algorithms
- How to create your own modified MuJoCo environment (Multi-task modifications can be pull-requested into gym-extensions)
How to run examples
Run TRPO
cd code; source activate rllab3; python run_trpo.py Hopper-v1
Run DDPG
cd code; source activate rllab3; python run_ddpg.py Hopper-v1
Plotting Results
cd code; python plot_results.py data/progress.csv Hopper-v1 --labels "trpo"
Manual testing of an env and custom env
cd code; python test_manual Hopper-v1
cd code; python test_modified_hopper_env_manually.py