org.encog.neural.networks.training.pnn

Class TrainBasicPNN

java.lang.Object

org.encog.ml.train.BasicTraining

org.encog.neural.networks.training.pnn.TrainBasicPNN

All Implemented Interfaces:

MLTrain, CalculationCriteria

public class TrainBasicPNN
extends BasicTraining
implements CalculationCriteria

Train a PNN.

Field Summary

Fields

Modifier and Type	Field and Description
static double	DEFAULT_MAX_ERROR The default max error.
static double	DEFAULT_MIN_IMPROVEMENT The default minimum improvement before stop.
static int	DEFAULT_NUM_SIGMAS The default number of sigmas to evaluate between the low and high.
static double	DEFAULT_SIGMA_HIGH The default sigma high value.
static double	DEFAULT_SIGMA_LOW THe default sigma low value.

Constructor Summary

Constructors

Constructor and Description
TrainBasicPNN(BasicPNN network, MLDataSet training) Train a BasicPNN.

Method Summary

Modifier and Type	Method and Description
double	calcErrorWithMultipleSigma(double[] x, double[] der1, double[] der2, boolean der) Calculate the error with multiple sigmas.
double	calcErrorWithSingleSigma(double sig) Calculate the error using a common sigma.
double	calculateError(MLDataSet training, boolean deriv) Calculate the error for the entire training set.
boolean	canContinue()
MLData	computeDeriv(MLData input, MLData target) Compute the derivative for target data.
double	getMaxError()
MLMethod	getMethod() Get the current best machine learning method from the training.
double	getMinImprovement()
int	getNumSigmas()
double	getSigmaHigh()
double	getSigmaLow()
void	iteration() Perform one iteration of training.
TrainingContinuation	pause() Pause the training to continue later.
void	resume(TrainingContinuation state) Resume training.
void	setMaxError(double maxError)
void	setMinImprovement(double minImprovement)
void	setNumSigmas(int numSigmas)
void	setSigmaHigh(double sigmaHigh)
void	setSigmaLow(double sigmaLow)

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

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

Methods inherited from class java.lang.Object

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

Field Detail

DEFAULT_MAX_ERROR

public static final double DEFAULT_MAX_ERROR

The default max error.

See Also:

Constant Field Values

DEFAULT_MIN_IMPROVEMENT

public static final double DEFAULT_MIN_IMPROVEMENT

The default minimum improvement before stop.

See Also:

Constant Field Values

DEFAULT_SIGMA_LOW

public static final double DEFAULT_SIGMA_LOW

THe default sigma low value.

See Also:

Constant Field Values

DEFAULT_SIGMA_HIGH

public static final double DEFAULT_SIGMA_HIGH

The default sigma high value.

See Also:

Constant Field Values

DEFAULT_NUM_SIGMAS

public static final int DEFAULT_NUM_SIGMAS

The default number of sigmas to evaluate between the low and high.

See Also:

Constant Field Values

Constructor Detail

TrainBasicPNN

public TrainBasicPNN(BasicPNN network,
                     MLDataSet training)

Train a BasicPNN.

Parameters:

network - The network to train.

training - The training data.

Method Detail

calcErrorWithMultipleSigma

public double calcErrorWithMultipleSigma(double[] x,
                                         double[] der1,
                                         double[] der2,
                                         boolean der)

Calculate the error with multiple sigmas.

Specified by:

calcErrorWithMultipleSigma in interface CalculationCriteria

Parameters:

x - The data.

der1 - The first derivative.

der2 - The 2nd derivatives.

der - Calculate the derivative.

Returns:

The error.

calcErrorWithSingleSigma

public double calcErrorWithSingleSigma(double sig)

Calculate the error using a common sigma.

Specified by:

calcErrorWithSingleSigma in interface CalculationCriteria

Parameters:

sig - The sigma to use.

Returns:

The training error.

calculateError

public double calculateError(MLDataSet training,
                             boolean deriv)

Calculate the error for the entire training set.

Parameters:

training - Training set to use.

deriv - Should we find the derivative.

Returns:

The error.

canContinue

public boolean canContinue()

Specified by:

canContinue in interface MLTrain

Returns:

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

computeDeriv

public MLData computeDeriv(MLData input,
                           MLData target)

Compute the derivative for target data.

Parameters:

input - The input.

target - The target data.

Returns:

The output.

getMaxError

public double getMaxError()

Returns:

the maxError

getMethod

public MLMethod getMethod()

Get the current best machine learning method from the training.

Specified by:

getMethod in interface MLTrain

Returns:

The best machine learning method.

getMinImprovement

public double getMinImprovement()

Returns:

the minImprovement

getNumSigmas

public int getNumSigmas()

Returns:

the numSigmas

getSigmaHigh

public double getSigmaHigh()

Returns:

the sigmaHigh

getSigmaLow

public double getSigmaLow()

Returns:

the sigmaLow

iteration

public void iteration()

Perform one iteration of training.

Specified by:

iteration in interface MLTrain

pause

public TrainingContinuation pause()

Pause the training to continue later.

Specified by:

pause in interface MLTrain

Returns:

A training continuation object.

resume

public void resume(TrainingContinuation state)

Resume training.

Specified by:

resume in interface MLTrain

Parameters:

state - The training continuation object to use to continue.

setMaxError

public void setMaxError(double maxError)

Parameters:

maxError - the maxError to set

setMinImprovement

public void setMinImprovement(double minImprovement)

Parameters:

minImprovement - the minImprovement to set

setNumSigmas

public void setNumSigmas(int numSigmas)

Parameters:

numSigmas - the numSigmas to set

setSigmaHigh

public void setSigmaHigh(double sigmaHigh)

Parameters:

sigmaHigh - the sigmaHigh to set

setSigmaLow

public void setSigmaLow(double sigmaLow)

Parameters:

sigmaLow - the sigmaLow to set