org.encog.neural.networks.training.propagation
Class Propagation

java.lang.Object
  org.encog.neural.networks.training.BasicTraining
      org.encog.neural.networks.training.propagation.Propagation

All Implemented Interfaces:

Train

Direct Known Subclasses:

Backpropagation, ManhattanPropagation, ResilientPropagation, ScaledConjugateGradient

public abstract class Propagation
extends BasicTraining

Implements basic functionality that is needed by each of the propagation methods. The specifics of each of the propagation methods is implemented inside of the PropagationMethod interface implementors.

Author:

jheaton

Constructor Summary

Propagation(BasicNetwork network, NeuralDataSet training)
Construct a propagation object.

Method Summary

boolean	canContinue()
void	finishTraining() Should be called after training has completed and the iteration method will not be called any further.
FlatNetwork	getCurrentFlatNetwork()
TrainFlatNetwork	getFlatTraining()
BasicNetwork	getNetwork() Get the current best network from the training.
int	getNumThreads()
OpenCLTrainingProfile	getProfile()
boolean	isValidResume(TrainingContinuation state) Determine if this specified training continuation object is valid for this training method.
void	iteration() Perform one training iteration.
void	iteration(int count) Perform the specified number of training iterations.
TrainingContinuation	pause() Pause the training to continue later.
void	resume(TrainingContinuation state) Resume training.
void	setFlatTraining(TrainFlatNetwork flatTraining)
void	setNumThreads(int numThreads) Set the number of threads.

Methods inherited from class org.encog.neural.networks.training.BasicTraining

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

Methods inherited from class java.lang.Object

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

Constructor Detail

Propagation

public Propagation(BasicNetwork network,
                   NeuralDataSet training)

Construct a propagation object.

Parameters:

network - The network.

training - The training set.

Method Detail

canContinue

public boolean canContinue()

Returns:

True if this training can be 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 Train

Overrides:

finishTraining in class BasicTraining

getCurrentFlatNetwork

public FlatNetwork getCurrentFlatNetwork()

Returns:

the currentFlatNetwork

getFlatTraining

public TrainFlatNetwork getFlatTraining()

Returns:

the flatTraining

getNetwork

public BasicNetwork getNetwork()

Description copied from interface: Train

Get the current best network from the training.

Returns:

The network.

getNumThreads

public int getNumThreads()

Returns:

The number of threads.

getProfile

public OpenCLTrainingProfile getProfile()

Returns:

The OpenCL device to use, or null for the CPU.

isValidResume

public boolean isValidResume(TrainingContinuation state)

Determine if this specified training continuation object is valid for this training method.

Parameters:

state - The training continuation object to check.

Returns:

True if the continuation object is valid.

iteration

public void iteration()

Perform one training iteration.

iteration

public void iteration(int count)

Perform the specified number of training iterations. This can be more efficient than single training iterations. This is particularly true if you are training with a GPU.

Specified by:

iteration in interface Train

Overrides:

iteration in class BasicTraining

Parameters:

count - The number of training iterations.

pause

public TrainingContinuation pause()

Pause the training to continue later.

Returns:

A training continuation object.

resume

public void resume(TrainingContinuation state)

Resume training.

Parameters:

state - The training continuation object to use to continue.

setFlatTraining

public void setFlatTraining(TrainFlatNetwork flatTraining)

Parameters:

flatTraining - the flatTraining to set

setNumThreads

public void setNumThreads(int numThreads)

Set the number of threads. Specify zero to tell Encog to automatically determine the best number of threads for the processor. If OpenCL is used as the target device, then this value is not used.

Parameters:

numThreads - The number of threads.