README.md

---
library_name: stable-baselines3
tags:
- LunarLander-v2
- deep-reinforcement-learning
- reinforcement-learning
- stable-baselines3
model-index:
- name: ppo
  results:
  - task:
      type: reinforcement-learning
      name: reinforcement-learning
    dataset:
      name: LunarLander-v2
      type: LunarLander-v2
    metrics:
    - type: mean_reward
      value: 249.17 +/- 16.55
      name: mean_reward
      verified: false
---

# **ppo** Agent playing **LunarLander-v2**
This is a trained model of a **ppo** agent playing **LunarLander-v2**
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).

## Usage (with Stable-baselines3)

```python
# In Colab, install packages if required:
#     gymnasium[box2d]: Contains the LunarLander-v2 environment 🌛
#     stable-baselines3[extra]: The deep reinforcement learning library.
#     huggingface_sb3: Additional code for Stable-baselines3 to load and upload models from the Hugging Face 🤗 Hub.
!apt install swig cmake
!pip install gymnasium[box2d] stable_baselines3[extra] huggingface-sb3

import gymnasium as gym
from huggingface_sb3 import load_from_hub
from stable_baselines3 import PPO
from stable_baselines3.common.vec_env import DummyVecEnv
from stable_baselines3.common.evaluation import evaluate_policy
from stable_baselines3.common.monitor import Monitor

# Create the evaluation environment
env_id = "LunarLander-v2"
eval_env = Monitor(gym.make(env_id), filename="./video.mp4")

# Load saved agent
repo_id = "davidkh/ppo-LunarLander-v2"
filename = "ppo-LunarLander-v2.zip"
checkpoint = load_from_hub(repo_id, filename)
model = PPO.load(checkpoint, print_system_info=True)

# Evaluation
mean_reward, std_reward = evaluate_policy(model, eval_env, n_eval_episodes=10, deterministic=True)
print(f"mean_reward={mean_reward:.2f} +/- {std_reward}")


# Example use of the trained agent
observation, info = eval_env.reset()
for _ in range(1000):
    eval_env.render()
    action, _states = model.predict(observation, deterministic=True)
    observation, rewards, terminated, truncated, info = eval_env.step(action)
    if terminated or truncated:
      print("Environment is reset")
      observation, info = eval_env.reset() 

eval_env.close()

...
```