/// <summary>
/// Initializes a new instance of KMeans algorithm
/// </summary>
///
/// <param name="k">The number of clusters to divide input data.</param>
/// <param name="distance">The distance function to use. Default is to
/// use the <see cref="Accord.Math.Distance.SquareEuclidean(double[], double[])"/> distance.</param>
///
public KMeans(int k, Func <double[], double[], double> distance)
{
if (k <= 0)
{
throw new ArgumentOutOfRangeException("k");
}
if (distance == null)
{
throw new ArgumentNullException("distance");
}
this.distance = distance;
// To store centroids of the clusters
this.proportions = new double[k];
this.centroids = new double[k][];
this.covariances = new double[k][, ];
// Create the object-oriented structure to hold
// information about the k-means' clusters.
List <KMeansCluster> clusterList = new List <KMeansCluster>(k);
for (int i = 0; i < k; i++)
{
clusterList.Add(new KMeansCluster(this, i));
}
this.clusters = new KMeansClusterCollection(this, clusterList);
}
/// <summary>
/// Initializes a new instance of the KMeans algorithm
/// </summary>
///
/// <param name="k">The number of clusters to divide the input data into.</param>
/// <param name="distance">The distance function to use. Default is to
/// use the <see cref="Accord.Math.Distance.SquareEuclidean(double[], double[])"/> distance.</param>
///
public KMeans(int k, Func <double[], double[], double> distance)
{
if (k <= 0)
{
throw new ArgumentOutOfRangeException("k");
}
if (distance == null)
{
throw new ArgumentNullException("distance");
}
// Create the object-oriented structure to hold
// information about the k-means' clusters.
this.clusters = new KMeansClusterCollection(k, distance);
}
/// <summary>
/// Initializes a new instance of the KMeans algorithm
/// </summary>
///
/// <param name="k">The number of clusters to divide the input data into.</param>
/// <param name="distance">The distance function to use. Default is to
/// use the <see cref="Accord.Math.Distance.SquareEuclidean(double[], double[])"/> distance.</param>
///
public KMeans(int k, Func <double[], double[], double> distance)
{
if (k <= 0)
{
throw new ArgumentOutOfRangeException("k");
}
if (distance == null)
{
throw new ArgumentNullException("distance");
}
this.Tolerance = 1e-5;
this.ComputeInformation = true;
this.UseCentroidSeeding = true;
// Create the object-oriented structure to hold
// information about the k-means' clusters.
this.clusters = new KMeansClusterCollection(k, distance);
}
/// <summary>
/// Initializes a new instance of the KMeans algorithm
/// </summary>
///
/// <param name="k">The number of clusters to divide the input data into.</param>
/// <param name="distance">The distance function to use. Default is to use the
/// <see cref="Accord.Math.Distance.SquareEuclidean(double[], double[])"/> distance.</param>
///
public KMeans(int k, IDistance <double[]> distance)
{
if (k <= 0)
{
throw new ArgumentOutOfRangeException("k");
}
if (distance == null)
{
throw new ArgumentNullException("distance");
}
this.Tolerance = 1e-5;
this.ComputeCovariances = true;
this.ComputeError = true;
this.UseSeeding = Seeding.KMeansPlusPlus;
this.MaxIterations = 0;
// Create the object-oriented structure to hold
// information about the k-means' clusters.
this.clusters = new KMeansClusterCollection(k, distance);
}
public static object KCluster(double[,] Data, int NumClusters)
{
int nRows = Data.GetLength(0), nCols = Data.GetLength(1);
double[][] data = new double[nRows][];
for (int r = 0; r < nRows; ++r)
{
data[r] = new double[nCols];
for (int c = 0; c < nCols; ++c)
{
data[r][c] = Data[r, c];
}
}
// Unfortunately, you can't set the initial cluster centers manually here, which
// means that every time this function is called, it will return different clusters.
aml.KMeans km = new aml.KMeans(NumClusters);
aml.KMeansClusterCollection kcc = km.Learn(data);
int nClusters = kcc.Count;
double[,] ret = new double[nClusters, nCols];
// TODO - 19Apr20 - needs testing, this used to previously put the mean[c] into ret[x,c]?
int x = 0;
foreach (KMeansClusterCollection.KMeansCluster cc in kcc.Clusters.OrderBy(xx => xx.Centroid.Mean()))
{
for (int c = 0; c < nCols; ++c)
{
ret[x, c] = cc.Centroid.Mean();
}
x++;
}
return(ret);
}
/// <summary>
/// Initializes a new instance of the <see cref="KMeansCluster"/> class.
/// </summary>
///
/// <param name="owner">The owner collection.</param>
/// <param name="index">The cluster index.</param>
///
public KMeansCluster(KMeansClusterCollection owner, int index)
{
this.owner = owner;
this.index = index;
}
