com.imsl.datamining.neural
Class QuasiNewtonTrainer.GradObjective

java.lang.Object
  com.imsl.datamining.neural.QuasiNewtonTrainer.Objective
      com.imsl.datamining.neural.QuasiNewtonTrainer.GradObjective

All Implemented Interfaces:

MinUnconMultiVar.Function, MinUnconMultiVar.Gradient

Direct Known Subclasses:

QuasiNewtonTrainer.BlockGradObjective

Enclosing class:

QuasiNewtonTrainer

protected class QuasiNewtonTrainer.GradObjective
extends QuasiNewtonTrainer.Objective
implements MinUnconMultiVar.Gradient

The Objective class is passed to the optimizer.

Field Summary

protected int	nFunctionEvaluations
protected int	nObs
protected int	nY

Method Summary

void	gradient(double[] weights, double[] gradient) Public interface for the gradient of the multivariate function to be minimized.

Methods inherited from class com.imsl.datamining.neural.QuasiNewtonTrainer.Objective

f

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.imsl.math.MinUnconMultiVar.Function

f

Field Detail

nFunctionEvaluations

protected int nFunctionEvaluations

nObs

protected int nObs

nY

protected int nY

Method Detail

gradient

public void gradient(double[] weights,
                     double[] gradient)

Description copied from interface: MinUnconMultiVar.Gradient

Public interface for the gradient of the multivariate function to be minimized.

Specified by:

gradient in interface MinUnconMultiVar.Gradient

Parameters:

weights - a double array, the point at which the gradient of the function is to be evaluated

gradient - a double array, the value of the gradient of the function at x