org.joone.net
Class NeuralNetValidator

java.lang.Object
  extended by org.joone.net.NeuralNetValidator
All Implemented Interfaces:
java.lang.Runnable, java.util.EventListener, NeuralNetListener

public class NeuralNetValidator
extends java.lang.Object
implements java.lang.Runnable, NeuralNetListener

This class is useful to validate a neural network. It simply sets some parameters of the neural network passed as parameter and starts itself in a separated thread, notifying a listener when the validation step finishes.

Author:
pmarrone

Constructor Summary
NeuralNetValidator(NeuralNet nn)
           
 
Method Summary
 void addValidationListener(NeuralValidationListener newListener)
           
 void cicleTerminated(NeuralNetEvent e)
           
 void errorChanged(NeuralNetEvent e)
           
 void fireNetValidated()
           
 NeuralNet getNeuralNet()
          Gets the network to validate (or has been validated).
 void netStarted(NeuralNetEvent e)
           
 void netStopped(NeuralNetEvent e)
           
 void netStoppedError(NeuralNetEvent e, java.lang.String error)
           
 void run()
           
 void start()
          Starts the validation into a separated thread
 void useTrainingData(boolean anUse)
          By default the validator validates a neural network with validation data, however by calling this method before calling the start() method, one can decide if the network should be validated with validation data (the parameter anUse should be false) or by using the training data (the parameter anUse should be true).
protected  void validate()
           
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

NeuralNetValidator

public NeuralNetValidator(NeuralNet nn)
Method Detail

addValidationListener

public void addValidationListener(NeuralValidationListener newListener)

validate

protected void validate()

fireNetValidated

public void fireNetValidated()

useTrainingData

public void useTrainingData(boolean anUse)
By default the validator validates a neural network with validation data, however by calling this method before calling the start() method, one can decide if the network should be validated with validation data (the parameter anUse should be false) or by using the training data (the parameter anUse should be true).

Parameters:
anUse - true if we should use training data for validation, false if we should use the validation data for validation (default).

start

public void start()
Starts the validation into a separated thread


run

public void run()
Specified by:
run in interface java.lang.Runnable

netStopped

public void netStopped(NeuralNetEvent e)
Specified by:
netStopped in interface NeuralNetListener

cicleTerminated

public void cicleTerminated(NeuralNetEvent e)
Specified by:
cicleTerminated in interface NeuralNetListener

netStarted

public void netStarted(NeuralNetEvent e)
Specified by:
netStarted in interface NeuralNetListener

errorChanged

public void errorChanged(NeuralNetEvent e)
Specified by:
errorChanged in interface NeuralNetListener

netStoppedError

public void netStoppedError(NeuralNetEvent e,
                            java.lang.String error)
Specified by:
netStoppedError in interface NeuralNetListener

getNeuralNet

public NeuralNet getNeuralNet()
Gets the network to validate (or has been validated).

Returns:
the netork to validate (or the network that has been validated).


Submit Feedback to pmarrone@users.sourceforge.net