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: 251.74 +/- 17.34
            name: mean_reward
            verified: false

PPO Agent Playing LunarLander-v2

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

Hyperparameters

{'env_id': 'LunarLander-v2'
'seed': 1
'num_envs': 16
'pi_hidden_layers': (64
64)
'v_hidden_layers': (64
64)
'learning_rate': 0.0003
'anneal_lr': False
'total_timesteps': 1000000
'num_steps': 1024
'gamma': 0.999
'gae_lambda': 0.98
'num_minibatches': 256
'num_epochs': 4
'norm_advantage': False
'clip_coef': 0.2
'clip_vloss': False
'vf_coef': 1.0
'ent_coef': 0.01
'max_grad_norm': 0.5
'target_kl': None}