pyXO

This is an implementation of the classic 3x3 game of Tic-Tac-Toe and is then generalized to an NxN game for any N. The code implements various algorithms like MiniMax, MiniMax with Aplha-Beta pruning and Monte Carlo Tree Search for the bots to play the game.

Setup

Clone the repository using:

git clone https://github.com/97amarnathk/pyXO.git

Packages required for running the program are sys and pygame. If these modules are not present then install the modules using:

pip install sys pygame

Run

To start the game, simply go to the directory in the terminal and run the main.py file using python main.py. This will open the game of Tic-Tac-Toe where you can play using the mouse. The color of your turn depends on which player number you are: 1 or -1. 1 corresponds to Blue and -1 corresponds to Pink.

There are five kinds of agents that can play the game:

1. Random Agent (RandomAgent())
2. User (Player())
3. MiniMax Agent (MiniMaxAgent())
4. MiniMax Agent with Alpha-Beta pruning (AlphaBetaMiniMaxAgent())
5. Monte Carlo Tree Search Agent (MCTSAgent())

To change the type of agent, go to main.py and change the type of agent to the required one in the players dictionary on line 27 and then run the file.

To change the size of the board to NxN, change the argument passed in the initialization of an object of class Board on line 23 of main.py to N. For example, to play a game on a 5x5 board, change: board = Board(3) --> board = Board(5)