org.apache.commons.math3.ml.neuralnet.twod.util

Class SmoothedDataHistogram

java.lang.Object

org.apache.commons.math3.ml.neuralnet.twod.util.SmoothedDataHistogram

All Implemented Interfaces:

MapDataVisualization

public class SmoothedDataHistogram
extends java.lang.Object
implements MapDataVisualization

Visualization of high-dimensional data projection on a 2D-map. The method is described in Using Smoothed Data Histograms for Cluster Visualization in Self-Organizing Maps
by Elias Pampalk, Andreas Rauber and Dieter Merkl.

Since:

3.6

Field Summary

Fields

Modifier and Type	Field and Description
private DistanceMeasure	distance Distance.
private double	membershipNormalization Normalization factor.
private int	smoothingBins Smoothing parameter.

Constructor Summary

Constructors

Constructor and Description
SmoothedDataHistogram(int smoothingBins, DistanceMeasure distance)

Method Summary

Methods

Modifier and Type	Method and Description
double[][]	computeImage(NeuronSquareMesh2D map, java.lang.Iterable<double[]> data) Creates an image of the data metrics when represented by the map.

Methods inherited from class java.lang.Object

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

Field Detail

smoothingBins

private final int smoothingBins

Smoothing parameter.

distance

private final DistanceMeasure distance

Distance.

membershipNormalization

private final double membershipNormalization

Normalization factor.

Constructor Detail

SmoothedDataHistogram

public SmoothedDataHistogram(int smoothingBins,
                     DistanceMeasure distance)

Parameters:

smoothingBins - Number of bins.

distance - Distance.

Method Detail

computeImage

public double[][] computeImage(NeuronSquareMesh2D map,
                      java.lang.Iterable<double[]> data)

Creates an image of the data metrics when represented by the map.

Specified by:

computeImage in interface MapDataVisualization

Parameters:

map - Map.

data - Data.

Returns:

a 2D-array (in row major order) representing the metrics.

Throws:

NumberIsTooSmallException - if the size of the map is smaller than the number of smoothing bins.