org.neuroph.util.random

Class DistortRandomizer

java.lang.Object

org.neuroph.util.random.WeightsRandomizer

org.neuroph.util.random.DistortRandomizer

public class DistortRandomizer
extends WeightsRandomizer

This class provides distort randomization technique, which distorts existing weight values using specified distortion factor. Weights are distorted using following formula: newWeightValue = currentWeightValue + (distortionFactor - (random * distortionFactor * 2))

Author:

Zoran Sevarac

Field Summary

Fields inherited from class org.neuroph.util.random.WeightsRandomizer

randomGen

Constructor Summary

Constructors

Constructor and Description
DistortRandomizer(double distortionFactor) Create a new instance of DistortRandomizer with specified distortion factor

Method Summary

Modifier and Type	Method and Description
protected void	randomize(Neuron neuron) Iterate all layers, neurons and connection weight and apply distort randomization

Methods inherited from class org.neuroph.util.random.WeightsRandomizer

getRandomGen, nextRandomWeight, randomize, randomize

Methods inherited from class java.lang.Object

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

Constructor Detail

DistortRandomizer

public DistortRandomizer(double distortionFactor)

Create a new instance of DistortRandomizer with specified distortion factor

Parameters:

distortionFactor - amount to distort existing weights

Method Detail

randomize

protected void randomize(Neuron neuron)

Iterate all layers, neurons and connection weight and apply distort randomization

Overrides:

randomize in class WeightsRandomizer

Parameters:

neuron -