For anyone trying to learn or practice RL, here’s a repo with working PyTorch implementations of 17 RL algorithms including DQN, DQN-HER, Double DQN, REINFORCE, DDPG, DDPG-HER, PPO, SAC, SAC Discrete, A3C, A2C etc..

Let me know what you think!

https://github.com/p-christ/Deep-Reinforcement-Learning-Algorithms-with-PyTorch