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

public 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, PropagationMethod method, NeuralDataSet training)
Construct a propagation trainer.

Method Summary

void	backwardPass(NeuralData ideal) Calculate the error for the recognition just done.
java.util.List<PropagationLevel>	getLevels()
BasicNetwork	getNetwork() Get the current best neural network.
NeuralOutputHolder	getOutputHolder()
void	iteration() Perform one iteration of training.

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

addStrategy, getError, getStrategies, getTraining, postIteration, preIteration, setError, 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,
                   PropagationMethod method,
                   NeuralDataSet training)

Construct a propagation trainer.

Parameters:

network - The network to train.

method - The propagation method to use.

training - The training data to use.

Method Detail

backwardPass

public void backwardPass(NeuralData ideal)

Calculate the error for the recognition just done.

Parameters:

ideal - What the output neurons should have yielded.

getLevels

public java.util.List<PropagationLevel> getLevels()

Returns:

The propagation levels.

getNetwork

public BasicNetwork getNetwork()

Get the current best neural network.

Returns:

The current best neural network.

getOutputHolder

public NeuralOutputHolder getOutputHolder()

Returns:

The output holder being used.

iteration

public void iteration()

Perform one iteration of training. Note: if you get a StackOverflowError while training, then you have endless recurrent loops. Try inserting no trainable synapses on one side of the loop.