README.md

metadata

library_name: stable-baselines3
tags:
  - LunarLander-v2
  - deep-reinforcement-learning
  - reinforcement-learning
  - stable-baselines3
model-index:
  - name: PPO
    results:
      - metrics:
          - type: mean_reward
            value: 266.93 +/- 24.72
            name: mean_reward
        task:
          type: reinforcement-learning
          name: reinforcement-learning
        dataset:
          name: LunarLander-v2
          type: 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 gym
from stable_baselines3 import PPO
from stable_baselines3.common.evaluation import evaluate_policy
from stable_baselines3.common.env_util import make_vec_env

# Create a vectorized environment of 16 parallel environments
env = make_vec_env("LunarLander-v2", n_envs=16)

# Optimizaed Hyperparameters
model = PPO(
  "MlpPolicy",
  env=env,
  n_steps=655,
  batch_size=32,
  n_epochs=8,
  gamma=0.998,
  gae_lambda=0.98,
  ent_coef=0.01,
  verbose=1,
)

# Train it for 500,000 timesteps
model.learn(total_timesteps=int(5e6))

# Create a new environment for evaluation
eval_env = gym.make("LunarLander-v2")

# Evaluate the model with 10 evaluation episodes and deterministic=True
mean_reward, std_reward = evaluate_policy(
  model, eval_env, n_eval_episodes=10, deterministic=True
)

# Print the results
print(f"mean_reward={mean_reward:.2f} +/- {std_reward}")

#>>> mean_reward=254.56 +/- 18.45056958672337