org.encog.engine.network.train.prop
Class TrainFlatNetworkResilient

java.lang.Object
  org.encog.engine.network.train.prop.TrainFlatNetworkProp
      org.encog.engine.network.train.prop.TrainFlatNetworkResilient

All Implemented Interfaces:

TrainFlatNetwork

public class TrainFlatNetworkResilient
extends TrainFlatNetworkProp

Train a flat network using RPROP.

Field Summary

Fields inherited from class org.encog.engine.network.train.prop.TrainFlatNetworkProp

currentError, gradients, network

Constructor Summary

TrainFlatNetworkResilient(FlatNetwork flat, EngineDataSet trainingSet)
Tran a network using RPROP.

TrainFlatNetworkResilient(FlatNetwork network, EngineDataSet training, double zeroTolerance, double initialUpdate, double maxStep)
Construct a resilient trainer for flat networks.

Method Summary

double[]	getUpdateValues()
double	updateWeight(double[] gradients, double[] lastGradient, int index) Calculate the amount to change the weight by.

Methods inherited from class org.encog.engine.network.train.prop.TrainFlatNetworkProp

calculateGradients, finishTraining, getError, getIteration, getLastGradient, getNetwork, getNumThreads, getTraining, iteration, iteration, learn, learnLimited, report, setIteration, setNumThreads

Methods inherited from class java.lang.Object

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

Constructor Detail

TrainFlatNetworkResilient

public TrainFlatNetworkResilient(FlatNetwork network,
                                 EngineDataSet training,
                                 double zeroTolerance,
                                 double initialUpdate,
                                 double maxStep)

Construct a resilient trainer for flat networks.

Parameters:

network - The network to train.

training - The training data to use.

zeroTolerance - How close a number should be to zero to be counted as zero.

initialUpdate - The initial update value.

maxStep - The maximum step value.

TrainFlatNetworkResilient

public TrainFlatNetworkResilient(FlatNetwork flat,
                                 EngineDataSet trainingSet)

Tran a network using RPROP.

Parameters:

flat - The network to train.

trainingSet - The training data to use.

Method Detail

updateWeight

public double updateWeight(double[] gradients,
                           double[] lastGradient,
                           int index)

Calculate the amount to change the weight by.

Specified by:

updateWeight in class TrainFlatNetworkProp

Parameters:

gradients - The gradients.

lastGradient - The last gradients.

index - The index to update.

Returns:

The amount to change the weight by.

getUpdateValues

public double[] getUpdateValues()

Returns:

The RPROP update values.