- Python version 3.7+
-
Step 1: Download or clone the repository
-
Step 2: Create a virtual enviroment
- Run the command: python -m venv venv
- Activate the virtual enviroment, run command: venv\Scripts\activate
- Install dependencies, run command: pip install -r requirements.txt
-
Step 3: Run the game
- Run the command: python main.py || you'll have a Q-learning agent on the left racket and a SARSA agent on the right racket
- Run the command: python main.py -q || you'll have a Q-learning agent on both racket
- Run the command: python main.py -sarsa || you'll have a SARSA agent on both racket
-
Step 1: Choose type of the model
- Consider file learning_q_sarsa.py if you want train a model with a Q-learning agent and a SARSA agent
- Consider file learning_q.py if you want train a model with two Q-learning agents
- Consider file learning_sarsa.py if you want train a model with SARSA agents
-
Step 2: Execute training of the model
- In learning file you can setup hyper-parameters(alpha, gamma and epsilon values), game speed, number of episodes and other configurations
- Execute your leaning file choosed in step 1, ex. run command: python learning_q_sarsa.py
-
Step 3: Load your model and play the game
- Run the command: python main.py || if you choose learning_q_sarsa.py file
- Run the command: python main.py -q || if you choose learning_q.py file
- Run the command: python main.py -sarsa || if you choose learning_sarsa.py file