MohanadMetrash/ppo-LunarLander-v3

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)

import gymnasium as gym
from stable_baselines3 import PPO
from huggingface_sb3 import load_from_hub

# --- Load the model from the Hugging Face Hub ---

# The repo_id is the name of the model repository on the Hub
repo_id = "MohanadMetrash/ppo-LunarLander-v3"
# The filename is the name of the saved model zip file
filename = "ppo-LunarLander-v3.zip"

# Download the model from the Hub and load it into a PPO object
checkpoint = load_from_hub(repo_id, filename)
model = PPO.load(checkpoint)


# --- Use the trained agent ---

# Create a new environment for the agent to play in
# "human" render_mode opens a window to display the environment
env = gym.make("LunarLander-v3", render_mode="human")

# Reset the environment to get the initial observation
obs, info = env.reset()

# Run the agent in the environment for 1000 steps
for _ in range(1000):
    # The model predicts the best action to take based on the observation
    action, _states = model.predict(obs, deterministic=True)
    
    # The environment takes the action and returns the next observation, reward, etc.
    obs, reward, terminated, truncated, info = env.step(action)
    
    # If the episode is finished (the lander has landed or crashed), reset the environment
    if terminated or truncated:
        print("Episode finished. Resetting environment.")
        obs, info = env.reset()

env.close()