org.encog.neural.networks.structure
Class AnalyzeNetwork

java.lang.Object
  extended by org.encog.neural.networks.structure.AnalyzeNetwork

public class AnalyzeNetwork
extends Object

Allows the weights and bias values of the neural network to be analyzed.


Constructor Summary
AnalyzeNetwork(BasicNetwork network)
          Construct a network analyze class.
 
Method Summary
 double[] getAllValues()
           
 NumericRange getBias()
           
 double[] getBiasValues()
           
 int getDisabledConnections()
           
 int getTotalConnections()
           
 NumericRange getWeights()
           
 NumericRange getWeightsAndBias()
           
 double[] getWeightValues()
           
 String toString()
          
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait
 

Constructor Detail

AnalyzeNetwork

public AnalyzeNetwork(BasicNetwork network)
Construct a network analyze class. Analyze the specified network.

Parameters:
network - The network to analyze.
Method Detail

getAllValues

public final double[] getAllValues()
Returns:
All of the values in the neural network.

getBias

public final NumericRange getBias()
Returns:
The numeric range of the bias values.

getBiasValues

public final double[] getBiasValues()
Returns:
The bias values in the neural network.

getDisabledConnections

public final int getDisabledConnections()
Returns:
The number of disabled connections in the network.

getTotalConnections

public final int getTotalConnections()
Returns:
The total number of connections in the network.

getWeights

public final NumericRange getWeights()
Returns:
The numeric range of the weights values.

getWeightsAndBias

public final NumericRange getWeightsAndBias()
Returns:
The numeric range of the weights and bias values.

getWeightValues

public final double[] getWeightValues()
Returns:
The weight values in the neural network.

toString

public final String toString()

Overrides:
toString in class Object


Copyright © 2014. All Rights Reserved.