- ./research/slitherin/: Slitherin' from OpenAI Request for Research 2.0 (computer self-play of classic snake game)
- ./gym-snake: classic snake game environment
- ./reinforce.py is a quick replication of a classic RL algorithm: REINFORCE
To run reinforce.py Example use:
python reinforce.py CartPole-v0 -n 100 -b 1000 --discount .9 -num_exp 1 --animate --seed 16 --use_reward_to_go -bl
To use tensorboard: The ./output directory is created within reinforce.py. Call:
tensorboard --logdir ./output/<Experiment path>