README.md

---
license: mit
pipeline_tag: reinforcement-learning
library_name: ml-agents
tags:
- BVS.1
- BVS
- deep-reinforcement-learning
- ML-Agents
---
# ppo Agent playing BVS V1


This is a trained model of a ppo agent playing [BroVSister](https://huggingface.co/spaces/RafaelJaime/BroVSister) using the [Godot rl agents](https://github.com/edbeeching/godot_rl_agents) using the example of stable_baselines3.

## Resume the training

  ```bash
  python stable_baselines3_example.py --timesteps=100_000 --onnx_export_path=model.onnx 
--save_model_path=model.zip --save_checkpoint_frequency=20_000 --experiment_name=exp1
  ```

### Watch your Agent play

  You can watch your agent **playing directly in your browser**

  1. Go to [BVS Huggingface space](https://huggingface.co/spaces/RafaelJaime/BroVSister) and go to character select menu
  2. Step 1: Find your model_id: RafaelJaime/*yourBVSmodelname*
  3. Step 2: Select your *.nn /*.onnx file
  4. Click on Watch the agent play 👀