Simple ppo implementation adapted to use random network distillation (RND). PyTorch implementation has RND for contienous action spaces with future support for discrete.
cd PPO/Torch
python main.py
- [] Seperate PPO and RND
- [] Support Discrete and Contineous Envs
- [] Vectorize PPO