org.encog.neural.networks.training.propagation.sgd

Class StochasticGradientDescent

java.lang.Object

org.encog.ml.train.BasicTraining

org.encog.neural.networks.training.propagation.sgd.StochasticGradientDescent

All Implemented Interfaces:

MLTrain, LearningRate, Momentum

public class StochasticGradientDescent
extends BasicTraining
implements Momentum, LearningRate

Constructor Summary

Constructors

Constructor and Description
StochasticGradientDescent(ContainsFlat network, MLDataSet training)
StochasticGradientDescent(ContainsFlat network, MLDataSet training, GenerateRandom theRandom)

Method Summary

Modifier and Type	Method and Description
void	calculateRegularizationPenalty(double[] l)
boolean	canContinue()
int	getBatchSize()
FlatNetwork	getFlat()
double	getL1()
double	getL2()
double	getLearningRate()
MLMethod	getMethod() Get the current best machine learning method from the training.
double	getMomentum()
UpdateRule	getUpdateRule()
boolean	isValidResume(TrainingContinuation state)
void	iteration() Perform one iteration of training.
void	layerRegularizationPenalty(int fromLayer, double[] l)
TrainingContinuation	pause() Pause the training.
void	preIteration() Call the strategies before an iteration.
void	process(MLDataPair pair)
void	resetError()
void	resume(TrainingContinuation state) Resume training.
void	setBatchSize(int theBatchSize)
void	setL1(double l1)
void	setL2(double l2)
void	setLearningRate(double rate) Set the learning rate.
void	setMomentum(double m) Set the momentum.
void	setUpdateRule(UpdateRule updateRule)
void	update()

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

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

Methods inherited from class java.lang.Object

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

Constructor Detail

StochasticGradientDescent

public StochasticGradientDescent(ContainsFlat network,
                                 MLDataSet training)

StochasticGradientDescent

public StochasticGradientDescent(ContainsFlat network,
                                 MLDataSet training,
                                 GenerateRandom theRandom)

Method Detail

process

public void process(MLDataPair pair)

update

public void update()

resetError

public void resetError()

iteration

public void iteration()

Description copied from interface: MLTrain

Perform one iteration of training.

Specified by:

iteration in interface MLTrain

canContinue

public boolean canContinue()

Specified by:

canContinue in interface MLTrain

Returns:

True if the training can be paused, and later continued.

getLearningRate

public double getLearningRate()

Specified by:

getLearningRate in interface LearningRate

Returns:

The learning rate.

getMomentum

public double getMomentum()

Specified by:

getMomentum in interface Momentum

Returns:

The momentum.

isValidResume

public boolean isValidResume(TrainingContinuation state)

pause

public TrainingContinuation pause()

Pause the training.

Specified by:

pause in interface MLTrain

Returns:

A training continuation object to continue with.

resume

public void resume(TrainingContinuation state)

Description copied from interface: MLTrain

Resume training.

Specified by:

resume in interface MLTrain

Parameters:

state - The training continuation object to use to continue.

getMethod

public MLMethod getMethod()

Description copied from interface: MLTrain

Get the current best machine learning method from the training.

Specified by:

getMethod in interface MLTrain

Returns:

The best machine learning method.

setLearningRate

public void setLearningRate(double rate)

Description copied from interface: LearningRate

Set the learning rate.

Specified by:

setLearningRate in interface LearningRate

Parameters:

rate - The new learning rate

setMomentum

public void setMomentum(double m)

Description copied from interface: Momentum

Set the momentum.

Specified by:

setMomentum in interface Momentum

Parameters:

m - The new momentum.

preIteration

public void preIteration()

Description copied from class: BasicTraining

Call the strategies before an iteration.

Overrides:

preIteration in class BasicTraining

getBatchSize

public int getBatchSize()

setBatchSize

public void setBatchSize(int theBatchSize)

getL1

public double getL1()

setL1

public void setL1(double l1)

getL2

public double getL2()

setL2

public void setL2(double l2)

calculateRegularizationPenalty

public void calculateRegularizationPenalty(double[] l)

layerRegularizationPenalty

public void layerRegularizationPenalty(int fromLayer,
                                       double[] l)

getFlat

public FlatNetwork getFlat()

getUpdateRule

public UpdateRule getUpdateRule()

setUpdateRule

public void setUpdateRule(UpdateRule updateRule)