minimax函数似乎没有输出正确的评估板

Question

我正在使用Python用MiniMax算法创建一个无与伦比的TicTacToe游戏。我的代码似乎没有输出每个位置的正确评估。根据我的理解，算法应该通过假设人类玩家最佳地发挥作用来“测试”每个位置。但是，评估始终返回NoneType。

我尝试在评估函数中使用其他数字。但是，这样做似乎没有返回正确的评估，或者至少，我认为这是不正确的。

board= ["O", 1, "X", 
        "X", 4, "X", 
        6, "O", "O"]

human = "O"
robot= "X"


#winning board combinations
def winning(board, player):
        #horizontal tests
    if((board[0] == player and board[1]==player and board[2]==player) or
        (board[3]==player and board[4]==player and board[5]==player) or
        (board[6]==player and board[7]==player and board[8]==player) or
        #diagonal tests
        (board[0]==player and board[4]==player and board[8]==player) or
        (board[2]==player and board[4]==player and board[6]==player) or
        #vertical tests
        (board[0]==player and board[3]==player and board[6]==player) or
        (board[1]==player and board[4]==player and board[7]==player) or
        (board[2]==player and board[5]==player and board[8]==player)
        ):
        return True
    else:
        return False


#evaluates the score of the game
def evaluation(board):
    if (winning(board, human)):
        return +10
    elif (winning(board, robot)):
        return -10
    elif(len(empty_spots)==0):
        return 0
    #trying to get another value
    # else:
        # return 2


#this is to find the empty spots on the board 
def empty_spots_func(board):
    for i in board:
        if(i!="X" and i!="O"):
            empty_spots.append(i)


def minimax(spot, empty_spots, depth, maximizing_player):
    if depth<=0:
        eval= evaluation(board)
        print(f'The evaluation function returns: {evaluation(board)}')
        return eval

    if maximizing_player:
        maxEval= -1000000
        #for each move in all possible moves
        for spot in empty_spots:

            #make the move
            board[spot]=robot

            #evaluate the outcome of the move
            eval = minimax(spot, empty_spots, depth-1, False)

            #then remove the move and replace it with the number
            board[spot]=spot

            maxEval= max(maxEval, eval)

        # print(f'The maximum evaluation {maxEval}')
        return maxEval

    else:
        minEval= +1000000
        #for each move in all possible moves
        for spot in empty_spots:

            #make the move 
            board[spot]=human

            #evaluate the outcome of the move
            eval= minimax(spot, empty_spots, depth-1, True)

            #then remove the spot
            board[spot]=spot

            #figure out the minimal evaluation
            minEval=min(minEval, eval)

        # print(f'The minimal evaluation is {minEval}')
        return minEval

现在在主函数中测试一个位置：

    test_spot_eval=minimax(0, empty_spots, len(empty_spots), True)
    empty_spots_func(board)
    print(f'The spot eval is {test_spot_eval}')

如果没有评估中的最后一个“else”，minimax只返回一个NoneType。

Answer 1

我制作了一个单独的程序来检查评估和获胜组合。事实证明，我所犯的错误是：

1. 在接受用户输入后，我没有将人工输入放在电路板上
2. 我没有注意到我错误地使评估函数返回了人类胜利的正值，而机器人胜利的负值则应该反过来。

可能还有更多我还没有注意到的错误。

错误1导致我的评估函数返回NoneType，因为在没有人工输入的情况下，算法无法评估获胜位置。