kris072008/ppo-LunarLander-v2

PPO Agent Playing LunarLander

A PPO agent written from scratch in PyTorch (CleanRL-style) for Unit 8 of the Hugging Face Deep RL Course. Trained on gymnasium LunarLander-v3 (the maintained version of LunarLander-v2).

Hyperparameters

{"seed": 1, "num_envs": 16, "num_steps": 1024, "total_timesteps": 5000000, "learning_rate": 0.0003, "gamma": 0.999, "gae_lambda": 0.98, "num_minibatches": 4, "update_epochs": 4, "clip_coef": 0.2, "ent_coef": 0.01, "vf_coef": 0.5}