This project is an implementation of a classical Tower Defence game, which allows you to play it on your own, and it also includes a gym
-like environment for training RL agents.
- PyGame Tower Defence
- Tower Defence as gym environment
- DQN - IN PROGRESS
- inference script for DQN
- Gameplay recording for Immitation learning
- Clone repository
git clone https://github.com/KyloRen1/TowerDefenceRLagent
- Create python environment
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
- Run game
python -m src.main_game --cfg configs/game_config.yml
- Run agent training
python -m src.main_agent --game-cfg configs/game_config.yml \
--agent-cfg configs/agent_config.yml --world-speed 10