Quixo Python

Implementation of turn based game Quixo in Python. Minimax with alpha beta pruning algorithm is used for computer play.

Playing

Pass any player instance from players.py to play.py

Random vs Random

# 100 Iterations
# play.py
play(RandomPlayer(),RandomPlayer(),100)

AI vs Random

# 100 Iterations
# play.py
play(AI_Player(GOOD_AI_PLAYER),RandomPlayer(),100)

Moves

play(piece,move)
piece: tuple = (row,col)
move: int = Direction

Directions

SLIDE_LEFT = 0
SLIDE_UP = 1
SLIDE_RIGHT = 2
SLIDE_DOWN = 3

Illegal moves:

Only blank and self pieces can be moved.
Only edge pieces can be moved.

if row is 0 cant slide down
if row is board_size cant slide up
if col is 0 cant slide right
if col is board_size cant slide down

State

Piece Values

0 = Blank

1 = X

2 = O

Turn

1 = X

2 = O

Value Function

V = evaluate(board)

Game Terminal Values

V = 100 + depth (X wins with minimum number of movements)
V = - (100 + depth) (O wins with minimum number of movements)

Depth Reach Values

More owned pieces on the board is usually better in initial game states.
V = (X Piece Count) - (O Piece Count)

Board Methods

printBoard() - print game board
printTurn() - prints current turn
getPossibleMoves() - returns all possible moves for the current board.
Returns moves in list of (piece,move)
getPossiblePieceMove() returns all possible moves for a single piece for current board.
Returns (move)
isGameEnd() - Returns winner integer value if game ended else returns None
changeTurn() - Changes turn
shift() - Shifts board when moving a piece
Used by play() method
checkMove() - Check a move if its valid
Returns True or False