我试图在我的程序中获得Negmax算法。我明白了,但无论如何我对算法也有点怀疑。但是当currentDepth达到深度的最大值后,它必须停止这个循环。请帮帮我解决这个问题?
换句话说,如果当前深度与最大深度相比较大,则必须停止循环。但这不会发生在这里,请帮我停止这个循环。提前致谢。
但是,这是问题所在,我也提出了一些方法,但它没有用,因此,我删除了并将其作为评论放入了我的实验。这是我的代码的一部分:
if(currentGameDepth == maximumGameDepth)
{
// I included some methods, but it won't execute to out of the loop
// I tried to return Mathf.Infinite(), but it is not work..
// please help me
//-----------------------------------------------------------------
}
完整代码如下: - 在这里你可以看到,两个地方添加NegamaxAlgorithm
关键字:Negmax算法在这里: -
private static float NegamaxAlgorithm(Piece game, float maximumGameDepth, float currentGameDepth, GamesModel moves)
{
if(currentGameDepth == maximumGameDepth)
{
// I included some methods, but it won't execute to out of the loop
// I tried to return Mathf.Infinite(), but it is not work..
// please help me
//-----------------------------------------------------------------
}
bestGameMove = null;
float bestGameScore = Mathf.NegativeInfinity;
foreach (GamesModel m in MovesList1)
{
Moves move = Piece.AccordaingToEvaluationWhiteSide(m,game);
float recursedGameScore;
Moves currentGameMove = move;
recursedGameScore = NegamaxAlgorithm(game , maximumGameDepth , currentGameDepth + 1, currentGameMove);
float currentGameScore = -recursedGameScore;
if (currentGameScore > bestGameScore)
{
bestGameScore = currentGameScore;
bestGameMove = m;
}
}
return bestGameScore;
}
..
问题可能是您使用浮点数来表示整数深度,并且由于浮点精度问题,==
比较失败。
尝试类似的东西:
private static float NegamaxAlgorithm(Piece game,
int maximumGameDepth, int currentGameDepth, GamesModel moves)
{
if(currentGameDepth >= maximumGameDepth)
{
// terminate recursion
return float.MinValue;
}
else
{
// ... use recursion to calculate next level ...
return bestGameScore;
}
}