jhpro.nnet

Class KohonenFeatureMap

java.lang.Object

jhpro.nnet.KohonenFeatureMap

All Implemented Interfaces:

java.io.Serializable

public class KohonenFeatureMap
extends java.lang.Object

KohonenFeatureMap. Also called Selforganizing Map (SOM). A feedforward / feedback type neural net. Built of an input layer thats neurons are connected with each neuron of another layer, called "feature map". The feature map can be one- or two-dimensional and each of its neurons is connected to all other neurons on the map. Mainly used for classification.

It is probably the most useful neural net type, if the learning process of the human brain shall be simulated. The "heart" of this type is the feature map, a neuron layer where neurons are organizing themselves according to certain input values. The type of this neural net is both feedforward (input layer to feature map) and feedback (feature map).

Number of neuron layers	1 input layer, 1 feature map
Input dimension	minimum: 1, maximum: 3
Number of neurons in feature map	minimum: 1*1, maximum: N*M
Number of input values	minimum: 1, maximum: N
Weight matrices	Automatically created and initialized. One matrix connects input layer and feature map. The other matrix is not of the WeightMatrix type and connects the neurons of the feature map among themselves.
Biases	Not used
One learning step includes	Selection of a random input value, finding the most activated map neuron, changing of the weights

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
KohonenFeatureMap() Declares the object kfm to be of type KohonenFeatureMap.

Method Summary

Modifier and Type	Method and Description
void	connectLayers(InputMatrix inputmatrix) Connects each input neuron (automatically created, depending on the dimension of the input matrix im) with each neuron of the feature map.
void	createMapLayer(int i) Creates the net's feature map with xSize*ySize map neurons.
void	createMapLayer(int i, int j) Creates the net's feature map with xSize*ySize map neurons.
void	decreaseActivationArea()
boolean	displayNow()
void	error(int i)
boolean	finishedLearning()
double	getActivationArea()
java.lang.String	getElapsedTime() Returns the time that elapsed since the learning process started.
double	getInitActivationArea() Returns the initial activation area of the feature map.
double	getInitLearningRate() Returns the initial learning rate of the net.
InputValue[]	getInputValues()
int	getLearningCycle() Returns the current learning cycle of the net.
double	getLearningRate() Returns the current learning rate of the net.
MapNeuron[]	getMapNeurons()
int	getMapSizeX() Returns the size of the feature map in x-dimension.
int	getMapSizeY() Returns the size of the feature map in y-dimension.
int	getMaxLearningCycles()
int	getNumberOfWeights() Returns the number of weights in the weight matrix that connects the input neurons with the neurons of the feature map.
double	getStopArea() Returns the final activation area of the feature map.
float[][]	getWeightValues() Returns all weight values of the weight matrix that connects the input neurons with the neurons of the feature map.
void	incLearningCycle()
void	learn() Performs one learning cycle.
void	resetTime()
void	setActivationArea(double d)
void	setDisplayStep(int i)
void	setInitActivationArea(double d) Sets the net's initial activation area to x.
void	setInitLearningRate(double d) Set initial learning rate.
void	setLearningRate(double d)
void	setMaxLearningCycles(int i) Sets the maximum number of learning cycles to x.
void	setStopArea(double d) Sets the net's final activation area to x.
double	square(double d)

Methods inherited from class java.lang.Object

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

Constructor Detail

KohonenFeatureMap

public KohonenFeatureMap()

Declares the object kfm to be of type KohonenFeatureMap.

Method Detail

createMapLayer

public void createMapLayer(int i)

Creates the net's feature map with xSize*ySize map neurons.

Parameters:

i -

createMapLayer

public void createMapLayer(int i,
                           int j)

Creates the net's feature map with xSize*ySize map neurons.

Parameters:

i -

j -

connectLayers

public void connectLayers(InputMatrix inputmatrix)

Connects each input neuron (automatically created, depending on the dimension of the input matrix im) with each neuron of the feature map. Besides, all map neurons are connected among themselves. All weight matrices are created and initialized with random values, taken from the input matrix im.

Parameters:

inputmatrix -

setInitLearningRate

public void setInitLearningRate(double d)

Set initial learning rate. Sets the net's initial learning rate to x. The default value is 0.6.

Parameters:

d -

getInitLearningRate

public double getInitLearningRate()

Returns the initial learning rate of the net.

Returns:

setInitActivationArea

public void setInitActivationArea(double d)

Sets the net's initial activation area to x. The default value is the greater of both map sizes divided by 2.

Parameters:

d -

getInitActivationArea

public double getInitActivationArea()

Returns the initial activation area of the feature map.

Returns:

setActivationArea

public void setActivationArea(double d)

setStopArea

public void setStopArea(double d)

Sets the net's final activation area to x. The default value is initActivationArea divided by 10.

Parameters:

d -

getStopArea

public double getStopArea()

Returns the final activation area of the feature map.

Returns:

getActivationArea

public double getActivationArea()

getMapSizeX

public int getMapSizeX()

Returns the size of the feature map in x-dimension.

Returns:

getMapSizeY

public int getMapSizeY()

Returns the size of the feature map in y-dimension.

Returns:

getNumberOfWeights

public int getNumberOfWeights()

Returns the number of weights in the weight matrix that connects the input neurons with the neurons of the feature map.

Returns:

getWeightValues

public float[][] getWeightValues()

Returns all weight values of the weight matrix that connects the input neurons with the neurons of the feature map.

Returns:

getMapNeurons

public MapNeuron[] getMapNeurons()

getInputValues

public InputValue[] getInputValues()

decreaseActivationArea

public void decreaseActivationArea()

learn

public void learn()

Performs one learning cycle. This method is usually called within a loop, which exits, if the finishedLearning() method returns true.

finishedLearning

public boolean finishedLearning()

error

public void error(int i)

square

public double square(double d)

setLearningRate

public void setLearningRate(double d)

getLearningRate

public double getLearningRate()

Returns the current learning rate of the net.

Returns:

setDisplayStep

public void setDisplayStep(int i)

displayNow

public boolean displayNow()

resetTime

public void resetTime()

getElapsedTime

public java.lang.String getElapsedTime()

Returns the time that elapsed since the learning process started.

Returns:

setMaxLearningCycles

public void setMaxLearningCycles(int i)

Sets the maximum number of learning cycles to x. The default value is -1 (no maximum).

Parameters:

i -

getMaxLearningCycles

public int getMaxLearningCycles()

incLearningCycle

public void incLearningCycle()

getLearningCycle

public int getLearningCycle()

Returns the current learning cycle of the net.

Returns: