README.md

metadata

tags:
  - LunarLander-v2
  - ppo
  - deep-reinforcement-learning
  - reinforcement-learning
  - custom-implementation
  - deep-rl-course
model-index:
  - name: PPO
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: LunarLander-v2
          type: LunarLander-v2
        metrics:
          - type: mean_reward
            value: '-134.81 +/- 132.27'
            name: mean_reward
            verified: false

PPO Agent Playing LunarLander-v2

This is a trained model of a PPO agent playing LunarLander-v2.

Hyperparameters

{'exp_name': 'ppo'
'raw_exp_version': None
'n_hidden_layers': 1
'd_hidden_layers': 64
'env_id': 'LunarLander-v2'
'n_envs': 4
'total_timesteps': 1000000
'batch_timesteps': 128
'n_mini_batches': 4
'n_epochs': 4
'lr': 0.0003
'alpha': 0.2
'gamma': 0.99
'lmbda': 0.95
'normalize_advantages': True
'coef_vf': 0.5
'coef_s': 0.01
'max_grad_norm': 1.0
'seed': 2546260713
'cuda': True
'deterministic': True
'repo_id': 'knight9114/ppo-LunarLander-v2-unit8.1'
'batch_size': 512
'mini_batch_size': 128
'exp_version': 'version_13'}