Skip to content

A Mancala AI Agent Using Minimax with Alpha-Beta Pruning

Notifications You must be signed in to change notification settings

mazy1998/MancalaAI

Repository files navigation

Mancal AI

Install

To launch this application make sure you are in a Python 3.x environment with p5 package installed.

  1. Installing Python 3.x
  2. Follow the installation guide for p5
  3. Clone the Repository
  4. Play Mancala

Mancala Rules

Mancala is a two player game in which the players have to move the stones on the board and try to capture as many stones they can in their tray. The player that has the most stones in their tray at the end of the game wins.

There exist many variations of the game and our implementation is based on the rules defined here

Setup

There are 12 slots on the board, with 4 pebbles in each at the initial state of the game. The two player trays at each end are initially empty, with the tray on the right side belonging to Player 1 and the tray on the left belonging to AI/Player 2.

Playing

The game starts by showing the following screen to the user

Splash Screen

There are two modes available

  • 2 Player Mode: Play against another human
  • AI Mode: Play against our AI

This is the initial game state

Splash Screen

Algorithms

  • Minimax with Alpha-Beta Pruning
  • Alpha Beta Pruning
  • Neural Networks

About

A Mancala AI Agent Using Minimax with Alpha-Beta Pruning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published