org.nd4j.linalg.solvers

Class VectorizedDeepLearningGradientAscent

java.lang.Object

org.nd4j.linalg.solvers.VectorizedDeepLearningGradientAscent

All Implemented Interfaces:

OptimizerMatrix

public class VectorizedDeepLearningGradientAscent
extends Object
implements OptimizerMatrix

Vectorized Stochastic Gradient Ascent

Author:

Adam Gibson

Constructor Summary

Constructors

Constructor and Description
VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function)
VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function, double initialStepSize)
VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function, double initialStepSize, IterationListener listener)
VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function, IterationListener listener)

Method Summary

Modifier and Type	Method and Description
double	getInitialStepSize()
VectorizedBackTrackLineSearch	getLineMaximizer()
OptimizableByGradientValueMatrix	getOptimizable()
double	getStpmax()
boolean	isConverged() Whether the algorithm is converged
boolean	optimize() Run optimize
boolean	optimize(int numIterations) Run optimize up to the specified number of epochs
void	setInitialStepSize(double initialStepSize)
void	setMaxIterations(int maxIterations) The default max number of iterations to run
void	setMaxStepSize(double v)
void	setStpmax(double stpmax)
void	setTolerance(double tolerance) Sets the tolerance in the convergence test: 2.0*|value-old_value| <= tolerance*(|value|+|old_value|+eps) Default value is 0.001.

Methods inherited from class java.lang.Object

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

Constructor Detail

VectorizedDeepLearningGradientAscent

public VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function,
                                            double initialStepSize)

VectorizedDeepLearningGradientAscent

public VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function,
                                            IterationListener listener)

VectorizedDeepLearningGradientAscent

public VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function,
                                            double initialStepSize,
                                            IterationListener listener)

VectorizedDeepLearningGradientAscent

public VectorizedDeepLearningGradientAscent(OptimizableByGradientValueMatrix function)

Method Detail

setMaxIterations

public void setMaxIterations(int maxIterations)

Description copied from interface: OptimizerMatrix

The default max number of iterations to run

Specified by:

setMaxIterations in interface OptimizerMatrix

getOptimizable

public OptimizableByGradientValueMatrix getOptimizable()

isConverged

public boolean isConverged()

Description copied from interface: OptimizerMatrix

Whether the algorithm is converged

Specified by:

isConverged in interface OptimizerMatrix

Returns:

true if the algorithm converged, false otherwise

getLineMaximizer

public VectorizedBackTrackLineSearch getLineMaximizer()

setTolerance

public void setTolerance(double tolerance)

Sets the tolerance in the convergence test: 2.0*|value-old_value| <= tolerance*(|value|+|old_value|+eps) Default value is 0.001.

Specified by:

setTolerance in interface OptimizerMatrix

Parameters:

tolerance - tolerance for convergence test

getInitialStepSize

public double getInitialStepSize()

setInitialStepSize

public void setInitialStepSize(double initialStepSize)

getStpmax

public double getStpmax()

setStpmax

public void setStpmax(double stpmax)

optimize

public boolean optimize()

Description copied from interface: OptimizerMatrix

Run optimize

Specified by:

optimize in interface OptimizerMatrix

Returns:

whether the algorithm converged properly

optimize

public boolean optimize(int numIterations)

Description copied from interface: OptimizerMatrix

Run optimize up to the specified number of epochs

Specified by:

optimize in interface OptimizerMatrix

Parameters:

numIterations - the max number of epochs to run

Returns:

whether the algorihtm converged properly

setMaxStepSize

public void setMaxStepSize(double v)