An AI based on reinforcement learning using SARSA to play Tic Tac Toe.
The project was to implement the algorithm of an AI capable of learning and playing on a scalable board. However, the graphical interface is still at a rudimentary stage; there's a lot to be improved in the future.
- You need Python 3.9 to run the program. You can download it from python.org.
pip
must be installed.
Git clone this repository :
git clone https://github.com/celinenguyentu/TicTacToeAI && cd TicTacToeAI
Create and activate a virtual environment :
python -m venv env
source env/bin/activate
Install dependencies :
pip install -r requirements.txt
Launch the program :
python TicTacToe.py
- Allow each player to start the game alternatively.
- Add a mode to train the AI in order to observe its improvement over time.
- Add an option to pick the board size, up to a ?x? board.
- Optimize and enhance the graphical interface.