org.encog.neural.flat.train.gradient
Class GradientWorkerCPU

java.lang.Object
  extended by org.encog.neural.flat.train.gradient.GradientWorkerCPU
All Implemented Interfaces:
FlatGradientWorker, EngineTask

public class GradientWorkerCPU
extends Object
implements FlatGradientWorker

Worker class for the mulithreaded training of flat networks.


Constructor Summary
GradientWorkerCPU(FlatNetwork theNetwork, TrainFlatNetworkProp theOwner, MLDataSet theTraining, int theLow, int theHigh)
          Construct a gradient worker.
 
Method Summary
 FlatNetwork getNetwork()
          
 double[] getWeights()
           
 void run()
          Perform the gradient calculation for the specified index range.
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

GradientWorkerCPU

public GradientWorkerCPU(FlatNetwork theNetwork,
                         TrainFlatNetworkProp theOwner,
                         MLDataSet theTraining,
                         int theLow,
                         int theHigh)
Construct a gradient worker.

Parameters:
theNetwork - The network to train.
theOwner - The owner that is doing the training.
theTraining - The training data.
theLow - The low index to use in the training data.
theHigh - The high index to use in the training data.
Method Detail

getNetwork

public final FlatNetwork getNetwork()

Specified by:
getNetwork in interface FlatGradientWorker
Returns:
The network being trained by this thread.

getWeights

public final double[] getWeights()
Specified by:
getWeights in interface FlatGradientWorker
Returns:
The weights for this network.

run

public final void run()
Perform the gradient calculation for the specified index range.

Specified by:
run in interface EngineTask


Copyright © 2011. All Rights Reserved.