---
license: mpl-2.0
tags:
- reinforcement-learning
- stable-baselines3
- enduro
- atari
- sb3
- ppo
- control
model-index:
- name: Enduro-v5-PPO
  parameters: 2.2M
  results:
    - task:
        type: reinforcement-learning
        name: Reinforcement Learning
      dataset:
        name: Enduro-v5
        type: gymnasium
      metrics:
        - type: mean_reward
          value: 599.54 +/- 131.49
---

# PPO Agent for Enduro-v5

This is a Proximal Policy Optimization (PPO) agent trained on the `Enduro-v5` environment using Stable Baselines 3.

## Hyperparameters

See `config.json` for details.

## Requirements

- **Python:** 3.10

### Dependencies

```txt
gymnasium==1.0.0
ale_py==0.10.1
gymnasium[atari]
torch==2.4.0
stable_baselines3==2.4.1
opencv-python==25.0.1
```

## How to Load

```python
from huggingface_hub import hf_hub_download
from stable_baselines3 import PPO

model_path = hf_hub_download(repo_id="lucasschott/Enduro-v5-PPO", filename="model.zip")
agent = PPO.load(model_path)
```