README.md

metadata

library_name: stable-baselines3
tags:
  - LunarLander-v3
  - deep-reinforcement-learning
  - reinforcement-learning
  - stable-baselines3
model-index:
  - name: PPO
    results:
      - task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: LunarLander-v3
          type: LunarLander-v3
        metrics:
          - type: mean_reward
            value: 279.56 +/- 18.47
            name: mean_reward
            verified: false

PPO Agent playing LunarLander-v3

This is a trained model of a PPO agent playing LunarLander-v3 using the stable-baselines3 library.

Usage (with Stable-baselines3)

from stable_baselines3 import PPO
from huggingface_sb3 import load_from_hub

model_filename = load_from_hub(your_repo_id, your_filename)

model = PPO.load(model_filename)

env = gym.make({env_id}, render_mode="human")
    obs, info = env.reset()

    done = False

    while not done:
        action, _states = model.predict(obs, deterministic=True)
        obs, rewards, done, truncated, info = env.step(action)
        env.render()

    env.close()