ppo-LunarLander-v2

PPO Agent playing LunarLander-v2

This is a trained model of a PPO agent playing LunarLander-v2 using the stable-baselines3 library.

Usage (with Stable-baselines3)

import gymnasium

from huggingface_sb3 import load_from_hub

from stable_baselines3 import PPO
from stable_baselines3.common.env_util import make_vec_env
from stable_baselines3.common.evaluation import evaluate_policy
from stable_baselines3.common.monitor import Monitor

checkpoint = load_from_hub("raymond-andrade/ppo-LunarLander-v2", "ppo-LunarLander-v2.zip")
model = PPO.load(checkpoint)

env = make_vec_env("LunarLander-v2", n_envs=1)

mean_reward, std_reward = evaluate_policy(
    model,
    env,
    n_eval_episodes=20,
    deterministic=True,
)

print(f"Mean reward = {mean_reward:.2f} +/- {std_reward:.2f}")

Training (with Stable-baselines3)

env = make_vec_env('LunarLander-v2', n_envs=32)

### Or use the following for vectorized envs
"""
from stable_baselines3.common.env_util import make_vec_env
from stable_baselines3.common.vec_env import SubprocVecEnv

env = make_vec_env(
    "LunarLander-v2",
    n_envs=32,                
    vec_env_cls=SubprocVecEnv
)
"""

model = PPO(
    "MlpPolicy",
    env,
    learning_rate=5e-4,
    n_steps=1024,
    batch_size=64,
    n_epochs=4,
    gamma=0.995,
    gae_lambda=0.98,
    ent_coef=0.01,
    verbose=1,
)

model.learn(total_timesteps=3000000)

model.save("your-desired-model-name")