def test_blank_board():
"""
Testing for result consistency between the 3 approaches to minimax algorithm
Testing algorithm result from evaluating a blank board
"""
board = create_board()
minimax_result = minimax(board, get_depth(board), True)
minimax_soft_alpha_beta_result = minimax_soft_alpha_beta(
board, get_depth(board), True, -inf, +inf)
minimax_alpha_beta_result = minimax_alpha_beta(board, get_depth(board),
True, -inf, +inf)
assert minimax_result == minimax_soft_alpha_beta_result
assert minimax_alpha_beta_result[0] == minimax_result[0]
assert minimax_alpha_beta_result[1][0] in minimax_result[1]
def test_board_bot_1st_turn():
"""
Testing for result consistency between the 3 approaches to minimax algorithm
Testing algorithm result from evaluating all possible states of the board where the bot starts first
"""
for turn_num in range(1, 9):
boards = get_all_possible_board_states(turn_num, BOT_STATE,
HUMAN_STATE)
is_maximizing_player = True if turn_num % 2 == 0 else False
for board in boards:
minimax_result = minimax(board, get_depth(board),
is_maximizing_player)
minimax_soft_alpha_beta_result = minimax_soft_alpha_beta(
board, get_depth(board), is_maximizing_player, -inf, +inf)
minimax_alpha_beta_result = minimax_alpha_beta(
board, get_depth(board), is_maximizing_player, -inf, +inf)
assert minimax_result == minimax_soft_alpha_beta_result
assert minimax_alpha_beta_result[0] == minimax_result[0]
assert minimax_alpha_beta_result[1][0] in minimax_result[1]
def make_move(self, board):
"""
Calls the minimax algorithm if the player is a bot, else request user input for human player.
:param board: type: numpy.ndarray
The current state of the Tic Tac Toe board game
Input for the minimax algorithm to find the optimal move
:return: type: tuple
Selected move index in numpy array format (<row_index>, <column_index>)
"""
if self._bot:
# Minimax algorithm
_, moves = minimax_soft_alpha_beta(board, get_depth(board), True,
-inf, +inf)
move = random.choice(moves)
else:
# Prompt the user to select a move
index = input("Enter move: ")
move = Player.convert_index_to_move(int(index))
return move
def bot_move_input_handler(board, bot):
"""
Calls the minimax algorithm and calculates the best possible move for the board state and updates the best possible
move on the board
:param board: type: numpy.array
The current state of the Tic Tac Toe board game
:param bot: type: class 'game.player.Player'
Player class instance of the bot player
"""
# Minimax algorithm
if all(box == BLANK_STATE for row in board for box in row):
# To reduce load, random the first turn for bot
row, box = (random.randint(-1, 2), random.randint(-1, 2))
else:
# Subsequent turn will not take long to calculate
_, moves = minimax_soft_alpha_beta(board, get_depth(board), True, -inf,
+inf)
row, box = random.choice(moves)
update_board(board, (row, box), bot)