将浮点数的二维数组转换为布尔值

Question

假设我有一个二维浮点数组

float [,] myfloats = new float[10, 10]
FillWithValues(myfloats);

如何使用随机评估方法将此二维浮点数组转换为二维布尔数组？

我希望是这样的：

bool[,] mybools;
mybools = myfloats.Select(myfloat => Evaluate(myfloat));

但这不起作用...

Answer 1

不幸的是，多维数组与 LINQ 不兼容并且有点难以处理。我认为没有办法绕过这个双循环解决方案：

bool[,] mybools = new bool[myfloats.GetLength(0), myfloats.GetLength(1)];
for (int x = 0; x<myfloats.GetLength(0); x++)
    for(int y=0; y<myfloats.GetLength(1); y++)
        mybools[x,y] = Evaluate(myfloats[x,y]);

---

Array.GetLength(Int32) 为您提供特定维度中数组的大小。

Answer 2

这不是特别有效，但有两种扩展方法可以做到这一点：

var floats = new float[3, 3] {
   { 1f, 2f, 3f },
   { 4f, 5f, 6f },
   { 7f, 8f, 9f }
};

Func<float, bool> floatToBool = f => f > 5f;

var bools =
   floats
   .ToJaggedArray()
   .Select(array => array.Select(floatToBool).ToArray())
   .ToArray()
   .To2DArray();

请注意，您必须将其操作回“数组数组”才能返回二维数组。

如果我们转换为 enumerable of enumerables 而不是锯齿状数组，效果会更好。然后它可能看起来像这样：

var bools =
   floats
   .ToEnumerableOfEnumerables()
   .Select(array => array.Select(floatToBool))
   .To2DArray();

第一部分的两个扩展方法是这样的：

[Pure, NotNull]
public static T[][] ToJaggedArray<T>([NotNull] this T[,] p2DArray) {
   var height = p2DArray.GetLength(0);
   var width = p2DArray.GetLength(1);

   var result = new T[height][];

   for (var row = 0; row < height; row++) {
      result[row] = new T[width];
      for (var col = 0; col < width; col++)
         result[row][col] = p2DArray[row, col];
   }

   return result;
}

[Pure, NotNull]
public static T[,] To2DArray<T>([NotNull] this T[][] pJaggedArray) {
   if (!pJaggedArray.Any() || !pJaggedArray[0].Any())
      return new T[0, 0];

   var height = pJaggedArray.Length;
   var width = pJaggedArray[0].Length;

   var result = new T[height, width];
   for (var r = 0; r < height; r++) {
      var subArray = pJaggedArray[r];
      if (subArray == null || subArray.Length != width)
         throw new InvalidOperationException("Jagged array was not rectangular.");

      for (var c = 0; c < width; c++)
         result[r, c] = subArray[c];
   }

   return result;
}

第二部分是这样的：

[Pure, NotNull]
public static IEnumerable<IEnumerable<T>> ToEnumerableOfEnumerables<T>(
   [NotNull] this T[,] p2DArray
) {
   var height = p2DArray.GetLength(0);

   for (var row = 0; row < height; row++)
      yield return p2DArray.GetPartialEnumerable(row);
}

[Pure, NotNull]
public static IEnumerable<T> GetPartialEnumerable<T>(
   [NotNull] this T[,] p2DArray,
   int pRow
) {
   var width = p2DArray.GetLength(1);

   for (var col = 0; col < width; col++)
      yield return p2DArray[pRow, col];
}

[Pure, NotNull]
public static T[,] To2DArray<T>(
   [NotNull] this IEnumerable<IEnumerable<T>> pJaggedArray
) {
   if (!pJaggedArray.Any() || !pJaggedArray.First().Any())
      return new T[0, 0];

   var height = pJaggedArray.Count();
   var width = pJaggedArray.First().Count();

   var result = new T[height, width];
   var r = 0;
   foreach (var subArray in pJaggedArray) {
      if (subArray == null || subArray.Count() != width)
         throw new InvalidOperationException("Jagged array was not rectangular.");

      var c = 0;
      foreach (var item in subArray)
         result[r, c++] = item;

      r += 1;
   }

   return result;
}