C# Accent K-最近邻函数参数问题

Question

我正在尝试使用 Accord 库的 K-最近邻函数，它可以与任何类型一起使用。

我的目标是将它与位图数据一起使用，但即使复制示例代码并粘贴它，我也会收到此错误：

Error   4   The best overloaded method match for 'Accord.MachineLearning.KNearestNeighbors<System.Drawing.Bitmap>.KNearestNeighbors(int, int, System.Drawing.Bitmap[], int[], Accord.Math.Distances.IDistance<System.Drawing.Bitmap>)' has some invalid arguments   D:\...\WindowsFormsApplication1\WidgetControl.cs    295 49  Project_Ochare

这真的很奇怪，因为它与示例完全相同。据我所知，参数应该可以工作。 我添加了 Accord 的主要、扩展核心、数学、数学扩展、数学代码、机器学习和统计作为参考。

我尝试搜索，但找不到任何答案..

示例代码与此相同： http://accord-framework.net/docs/html/T_Accord_MachineLearning_KNearestNeighbors_1.htm

怎么了？

这是他们出错的示例代码：

private void __Test()
{
    // The k-Nearest Neighbors algorithm can be used with
    // any kind of data. In this example, we will see how
   // it can be used to compare, for example, Strings.

    string[] inputs = 
    {
        "Car",    // class 0
        "Bar",    // class 0
        "Jar",    // class 0
        "Charm",  // class 1
        "Chair"   // class 1
    };
    int[] outputs =
    {
       0, 0, 0,  // First three are from class 0
       1, 1,     // And next two are from class 1
    };

    // Now we will create the K-Nearest Neighbors algorithm. For this
    // example, we will be choosing k = 1. This means that, for a given
    // instance, only its nearest neighbor will be used to cast a new
    // decision. 

    // In order to compare strings, we will be using Levenshtein's string distance
    KNearestNeighbors<string> knn = new KNearestNeighbors<string>(k: 1, classes: 2, inputs: inputs, outputs: outputs, distance: Distance.Levenshtein);

    // After the algorithm has been created, we can use it:
    int answer = knn.Compute("Chars"); // answer should be 1.
}

这些是我收到的错误消息。我用 KNearest 的 Accord 示例代码创建了一个新的空项目。 它要我将 Distance.Levenshtein 更改为 Distance.Levenshtein()， 然后它告诉我它需要参数，无论我如何编写或添加什么，它都会给出相同的错误。

例如 Distance.Levenshtein("", ""), Distance.Levenshtein(0, 0) Distance.Levenshtein("","")、Distance.Levenshtein(new string1, new string1) 等等……任何我能想到的尝试。

距离。提供了大量的测量功能，它们都会导致相同的错误。

错误信息：

Error   1   The best overloaded method match for 'Accord.MachineLearning.KNearestNeighbors<string>.KNearestNeighbors(int, int, string[], int[], Accord.Math.Distances.IDistance<string>)' has some invalid arguments    D:\Dropbox\C#\KNearestTest\KNearestTest\Form1.cs    48  45  KNearestTest


Error   2   Argument 5: cannot convert from 'method group' to 'Accord.Math.Distances.IDistance<string>' D:\Dropbox\C#\KNearestTest\KNearestTest\Form1.cs    49  59  KNearestTest

Answer 1

文档似乎与库不对应。 我建议编写自己的IDistance<string> 接口实现，使用Levenschtein 距离：

public class DistanceStringsLevenstein : IDistance<string>
{
    public double Distance(string x, string y)
    {
        return Accord.Math.Distance.Levenshtein(x, y);
    }
}

位图的实现是：

public class DistanceBitmapLevenstein : IDistance<Bitmap>
{
    public double Distance(Bitmap x, Bitmap y)
    {
        return Accord.Math.Distance.Levenshtein(ImageToByte(x), ImageToByte(y));
    }

    public static byte[] ImageToByte(Image img)
    {
        ImageConverter converter = new ImageConverter();
        return (byte[])converter.ConvertTo(img, typeof(byte[]));
    }
}

之后你可以在那个方法中使用这个类：

KNearestNeighbors<string> knn = new KNearestNeighbors<string>(k: 1, classes: 2,
        inputs: inputs, outputs: outputs, distance: new DistanceStringsLevenstein());