PPO Agent Playing LunarLander-v2
This is a trained model of a PPO agent playing LunarLander-v2.
Hyperparameters
{'checkpoint': 'models/LunarLander-v3__ppo__1__1768758426.pt'
'gym_id': 'LunarLander-v2'
'seed': 1
'num_episodes': 5
'capture_video': True
'cuda': True
'deterministic': True
'push_to_hub': True
'repo_id': 'BrennanDrake/ppo-LunarLander-v2-1'
'hub_commit_message': 'Add evaluation artifacts'
'hub_private': False}
Evaluation results
- mean_reward on LunarLander-v2self-reported288.27 +/- 18.16