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: '-147.57 +/- 73.26'
            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'
'anneal_lr': True
'clip_vloss': True
'gae': True
'gamma': 0.99
'ent_coeff': 0.01
'val_coeff': 0.5
'max_grad_norm': 0.5
'target_kl': None
'gae_lambda': 0.95
'learning_rate': 0.0001
'seed': 1993
'n_envs': 4
'total_timesteps': 25000
'n_timesteps': 32
'capture_video': False
'norm_adv': True
'clip_coef': 0.2
'mini_batch_size': 4
'epochs': 4}