smile.clustering

Class DeterministicAnnealing

java.lang.Object

smile.clustering.PartitionClustering

smile.clustering.CentroidClustering<double[],double[]>

smile.clustering.DeterministicAnnealing

All Implemented Interfaces:

java.io.Serializable, java.lang.Comparable<CentroidClustering<double[],double[]>>

public class DeterministicAnnealing
extends CentroidClustering<double[],double[]>

Deterministic annealing clustering. Deterministic annealing extends soft-clustering to an annealing process. For each temperature value, the algorithm iterates between the calculation of all posteriori probabilities and the update of the centroids vectors, until convergence is reached. The annealing starts with a high temperature. Here, all centroids vectors converge to the center of the pattern distribution (independent of their initial positions). Below a critical temperature the vectors start to split. Further decreasing the temperature leads to more splittings until all centroids vectors are separate. The annealing can therefore avoid (if it is sufficiently slow) the convergence to local minima.

References

1. Kenneth Rose. Deterministic Annealing for Clustering, Compression, Classification, Regression, and Speech Recognition.

See Also:

Serialized Form

Field Summary

Fields inherited from class smile.clustering.CentroidClustering

centroids, distortion

Fields inherited from class smile.clustering.PartitionClustering

k, OUTLIER, size, y

Constructor Summary

Constructors

Constructor and Description
DeterministicAnnealing(double distortion, double[][] centroids, int[] y) Constructor.

Method Summary

Modifier and Type	Method and Description
protected double	distance(double[] x, double[] y) The distance function.
static DeterministicAnnealing	fit(double[][] data, int Kmax) Clustering data into k clusters.
static DeterministicAnnealing	fit(double[][] data, int Kmax, double alpha, int maxIter, double tol, double splitTol) Clustering data into k clusters.

Methods inherited from class smile.clustering.CentroidClustering

compareTo, predict, toString

Methods inherited from class smile.clustering.PartitionClustering

run, seed

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

DeterministicAnnealing

public DeterministicAnnealing(double distortion,
                              double[][] centroids,
                              int[] y)

Constructor.

Parameters:

distortion - the total distortion.

centroids - the centroids of each cluster.

y - the cluster labels.

Method Detail

distance

protected double distance(double[] x,
                          double[] y)

Description copied from class: CentroidClustering

The distance function.

Specified by:

distance in class CentroidClustering<double[],double[]>

fit

public static DeterministicAnnealing fit(double[][] data,
                                         int Kmax)

Clustering data into k clusters.

Parameters:

data - the input data of which each row is an observation.

Kmax - the maximum number of clusters.

fit

public static DeterministicAnnealing fit(double[][] data,
                                         int Kmax,
                                         double alpha,
                                         int maxIter,
                                         double tol,
                                         double splitTol)

Clustering data into k clusters.

Parameters:

data - the input data of which each row is an observation.

Kmax - the maximum number of clusters.

alpha - the temperature T is decreasing as T = T * alpha. alpha has to be in (0, 1).

maxIter - the maximum number of iterations at each temperature.

tol - the tolerance of convergence test.

splitTol - the tolerance to split a cluster.