org.encog.neural.networks.training.propagation.manhattan
Class ManhattanPropagation

java.lang.Object
  org.encog.ml.train.BasicTraining
      org.encog.neural.networks.training.propagation.Propagation
          org.encog.neural.networks.training.propagation.manhattan.ManhattanPropagation

All Implemented Interfaces:

MLTrain, LearningRate, Train, MultiThreadable

public class ManhattanPropagation
extends Propagation
implements LearningRate

One problem that the backpropagation technique has is that the magnitude of the partial derivative may be calculated too large or too small. The Manhattan update algorithm attempts to solve this by using the partial derivative to only indicate the sign of the update to the weight matrix. The actual amount added or subtracted from the weight matrix is obtained from a simple constant. This constant must be adjusted based on the type of neural network being trained. In general, start with a higher constant and decrease it as needed. The Manhattan update algorithm can be thought of as a simplified version of the resilient algorithm. The resilient algorithm uses more complex techniques to determine the update value.

Author:

jheaton

Field Summary

Fields inherited from class org.encog.neural.networks.training.propagation.Propagation

gradients, lastError, network

Constructor Summary

ManhattanPropagation(ContainsFlat network, MLDataSet training, double theLearnRate)
Construct a Manhattan propagation training object.

Method Summary

boolean	canContinue() This training type does not support training continue.
double	getLearningRate()
void	initOthers() Perform training method specific init.
TrainingContinuation	pause() This training type does not support training continue.
void	resume(TrainingContinuation state) This training type does not support training continue.
void	setLearningRate(double rate) Set the learning rate.
double	updateWeight(double[] gradients, double[] lastGradient, int index) Calculate the amount to change the weight by.

Methods inherited from class org.encog.neural.networks.training.propagation.Propagation

calculateGradients, finishTraining, fixFlatSpot, getCurrentFlatNetwork, getLastGradient, getMethod, getThreadCount, iteration, iteration, learn, learnLimited, report, rollIteration, setErrorFunction, setThreadCount

Methods inherited from class org.encog.ml.train.BasicTraining

addStrategy, getError, getImplementationType, getIteration, getStrategies, getTraining, isTrainingDone, postIteration, preIteration, setError, setIteration, setTraining

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.encog.ml.train.MLTrain

addStrategy, getError, getImplementationType, getIteration, getStrategies, getTraining, isTrainingDone, setError, setIteration

Constructor Detail

ManhattanPropagation

public ManhattanPropagation(ContainsFlat network,
                            MLDataSet training,
                            double theLearnRate)

Construct a Manhattan propagation training object.

Parameters:

network - The network to train.

training - The training data to use.

theLearnRate - The learning rate.

Method Detail

getLearningRate

public final double getLearningRate()

Specified by:

getLearningRate in interface LearningRate

Returns:

The learning rate that was specified in the constructor.

setLearningRate

public final void setLearningRate(double rate)

Set the learning rate.

Specified by:

setLearningRate in interface LearningRate

Parameters:

rate - The new learning rate.

canContinue

public final boolean canContinue()

This training type does not support training continue.

Specified by:

canContinue in interface MLTrain

Returns:

Always returns false.

pause

public final TrainingContinuation pause()

This training type does not support training continue.

Specified by:

pause in interface MLTrain

Returns:

Always returns null.

resume

public final void resume(TrainingContinuation state)

This training type does not support training continue.

Specified by:

resume in interface MLTrain

Parameters:

state - Not used.

updateWeight

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

Calculate the amount to change the weight by.

Specified by:

updateWeight in class Propagation

Parameters:

gradients - The gradients.

lastGradient - The last gradients.

index - The index to update.

Returns:

The amount to change the weight by.

initOthers

public void initOthers()

Perform training method specific init.

Specified by:

initOthers in class Propagation