org.joone.engine

Class GaussianSpatialMap

java.lang.Object

org.joone.engine.SpatialMap

org.joone.engine.GaussianSpatialMap

All Implemented Interfaces:

java.io.Serializable

public class GaussianSpatialMap
extends SpatialMap

This class implements the SpatialMap interface providing a circular spatial map for use with the GaussianLayer and Kohonen Networks. The radius of the circle is equal to the initial Gaussian Size and is reduced if training is currently in process.

See Also:

Serialized Form

Field Summary

Fields inherited from class org.joone.engine.SpatialMap

TimeConstant

Constructor Summary

Constructors

Constructor and Description
GaussianSpatialMap() Creates a new instance of CircularSpatialMap

Method Summary

Methods

Modifier and Type	Method and Description
void	ApplyNeighborhoodFunction(double[] distances, double[] n_outs, boolean isLearning) Applies the neighborhood strategy based on this spatial maps implementation.

Methods inherited from class org.joone.engine.SpatialMap

distanceBetween, extractWinner, getCurrentGaussianSize, getInitialGaussianSize, getMapDepth, getMapHeight, getMapWidth, getOrderingPhase, getTimeConstant, getTotalEpochs, getWinnerX, getWinnerY, getWinnerZ, init, setCurrentGaussianSize, setInitialGaussianSize, setMapDepth, setMapDimensions, setMapHeight, setMapWidth, setOrderingPhase, setTimeConstant, updateCurrentGaussianSize

Methods inherited from class java.lang.Object

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

Constructor Detail

GaussianSpatialMap

public GaussianSpatialMap()

Creates a new instance of CircularSpatialMap

Method Detail

ApplyNeighborhoodFunction

public void ApplyNeighborhoodFunction(double[] distances,
                             double[] n_outs,
                             boolean isLearning)

Description copied from class: SpatialMap

Applies the neighborhood strategy based on this spatial maps implementation.

Specified by:

ApplyNeighborhoodFunction in class SpatialMap

Parameters:

distances - The euclidean distances between input and weights calculated by previous synapse.

n_outs - The outputs of this spatial maps neighborhood strategy.

isLearning - Is the network in the learning phase.