org.encog.neural.freeform.training

Class FreeformPropagationTraining

java.lang.Object

org.encog.ml.train.BasicTraining

org.encog.neural.freeform.training.FreeformPropagationTraining

All Implemented Interfaces:

java.io.Serializable, MLTrain

Direct Known Subclasses:

FreeformBackPropagation, FreeformResilientPropagation

public abstract class FreeformPropagationTraining
extends BasicTraining
implements java.io.Serializable

Provides basic propagation functions to other trainers.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
static double	FLAT_SPOT_CONST The constant to use to fix the flat spot problem.

Constructor Summary

Constructors

Constructor and Description
FreeformPropagationTraining() Don't use this constructor, it is for serialization only.
FreeformPropagationTraining(FreeformNetwork theNetwork, MLDataSet theTraining) Construct the trainer.

Method Summary

Modifier and Type	Method and Description
boolean	canContinue()
void	finishTraining() Should be called after training has completed and the iteration method will not be called any further.
int	getBatchSize()
double	getError()
TrainingImplementationType	getImplementationType()
int	getIteration()
MLMethod	getMethod() Get the current best machine learning method from the training.
MLDataSet	getTraining()
boolean	isFixFlatSopt()
void	iteration() Perform one iteration of training.
void	iteration(int count) Perform the specified number of training iterations.
void	setBatchSize(int batchSize) Set the batch size.
void	setError(double theError)
void	setFixFlatSopt(boolean fixFlatSopt) Set if we should fix the flat spot problem.
void	setIteration(int iteration) Set the current training iteration.

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

addStrategy, getStrategies, isTrainingDone, postIteration, preIteration, setTraining

Methods inherited from class java.lang.Object

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

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

pause, resume

Field Detail

FLAT_SPOT_CONST

public static final double FLAT_SPOT_CONST

The constant to use to fix the flat spot problem.

See Also:

Constant Field Values

Constructor Detail

FreeformPropagationTraining

public FreeformPropagationTraining()

Don't use this constructor, it is for serialization only.

FreeformPropagationTraining

public FreeformPropagationTraining(FreeformNetwork theNetwork,
                                   MLDataSet theTraining)

Construct the trainer.

Parameters:

theNetwork - The network to train.

theTraining - The training data.

Method Detail

canContinue

public boolean canContinue()

Specified by:

canContinue in interface MLTrain

Returns:

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

finishTraining

public void finishTraining()

Should be called after training has completed and the iteration method will not be called any further.

Specified by:

finishTraining in interface MLTrain

Overrides:

finishTraining in class BasicTraining

getError

public double getError()

Specified by:

getError in interface MLTrain

Overrides:

getError in class BasicTraining

Returns:

Returns the training error. This value is calculated as the training data is evaluated by the iteration function. This has two important ramifications. First, the value returned by getError() is meaningless prior to a call to iteration. Secondly, the error is calculated BEFORE training is applied by the call to iteration. The timing of the error calculation is done for performance reasons.

getImplementationType

public TrainingImplementationType getImplementationType()

Specified by:

getImplementationType in interface MLTrain

Overrides:

getImplementationType in class BasicTraining

Returns:

The training implementation type.

getIteration

public int getIteration()

Specified by:

getIteration in interface MLTrain

Overrides:

getIteration in class BasicTraining

Returns:

the iteration

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.

getTraining

public MLDataSet getTraining()

Specified by:

getTraining in interface MLTrain

Overrides:

getTraining in class BasicTraining

Returns:

The training data to use.

isFixFlatSopt

public boolean isFixFlatSopt()

Returns:

True, if we are fixing the flat spot problem.

iteration

public void iteration()

Perform one iteration of training.

Specified by:

iteration in interface MLTrain

iteration

public void iteration(int count)

Perform the specified number of training iterations. This is a basic implementation that just calls iteration the specified number of times. However, some training methods, particularly with the GPU, benefit greatly by calling with higher numbers than 1.

Specified by:

iteration in interface MLTrain

Overrides:

iteration in class BasicTraining

Parameters:

count - The number of training iterations.

setError

public void setError(double theError)

Specified by:

setError in interface MLTrain

Overrides:

setError in class BasicTraining

Parameters:

theError - Set the current error rate. This is usually used by training strategies.

setFixFlatSopt

public void setFixFlatSopt(boolean fixFlatSopt)

Set if we should fix the flat spot problem.

Parameters:

fixFlatSopt - True, if we should fix the flat spot problem.

setIteration

public void setIteration(int iteration)

Set the current training iteration.

Specified by:

setIteration in interface MLTrain

Overrides:

setIteration in class BasicTraining

Parameters:

iteration - the iteration to set

getBatchSize

public int getBatchSize()

Returns:

The batch size.

setBatchSize

public void setBatchSize(int batchSize)

Set the batch size.

Parameters:

batchSize - The batch size.