CharithAnupama/ppo-LunarLander-v2

PPO Agent Playing LunarLander-v2

Trained with a PPO implementation from scratch (CleanRL-style) in PyTorch.

Results

• Mean reward: 8.59
• Std reward: 73.34

Hyperparameters

• exp_name: ppo_from_scratch
• seed: 1
• cuda: 1
• env_id: LunarLander-v2
• total_timesteps: 400000
• learning_rate: 0.00025
• num_envs: 8
• num_steps: 128
• anneal_lr: 1
• gamma: 0.99
• gae_lambda: 0.95
• num_minibatches: 4
• update_epochs: 4
• clip_coef: 0.2
• ent_coef: 0.01
• vf_coef: 0.5
• max_grad_norm: 0.5
• repo_id: CharithAnupama/ppo-LunarLander-v2
• batch_size: 1024
• minibatch_size: 256