org.deeplearning4j.optimize.solvers

Class ConjugateGradient

java.lang.Object

org.deeplearning4j.optimize.solvers.BaseOptimizer

org.deeplearning4j.optimize.solvers.ConjugateGradient

All Implemented Interfaces:

Serializable, ConvexOptimizer

public class ConjugateGradient
extends BaseOptimizer

Modified based on cc.mallet.optimize.ConjugateGradient

no termination when zero tolerance

Since:

2013-08-25

Author:

Adam Gibson

See Also:

Serialized Form

Field Summary

Fields inherited from class org.deeplearning4j.optimize.solvers.BaseOptimizer

adaGrad, adaGradForVariable, conf, GRADIENT_KEY, iteration, iterationListeners, lineMaximizer, log, model, oldScore, PARAMS_KEY, score, SCORE_KEY, searchState, step, stepFunction, stpMax, terminationConditions

Constructor Summary

Constructors

Constructor and Description
ConjugateGradient(NeuralNetConfiguration conf, StepFunction stepFunction, Collection<IterationListener> iterationListeners, Collection<TerminationCondition> terminationConditions, Model model)
ConjugateGradient(NeuralNetConfiguration conf, StepFunction stepFunction, Collection<IterationListener> iterationListeners, Model model)

Method Summary

Modifier and Type	Method and Description
void	postStep() Post step (conjugate gradient among other methods needs this)
void	preProcessLine(org.nd4j.linalg.api.ndarray.INDArray line) Pre process the line (scaling and the like)
void	setupSearchState(Pair<Gradient,Double> pair) Setup the initial search state

Methods inherited from class org.deeplearning4j.optimize.solvers.BaseOptimizer

adaGradForVariables, batchSize, getAdaGrad, getAdaGradForVariable, gradientAndScore, optimize, postFirstStep, preFirstStepProcess, score, setBatchSize, updateGradientAccordingToParams, updateGradientAccordingToParams

Methods inherited from class java.lang.Object

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

Constructor Detail

ConjugateGradient

public ConjugateGradient(NeuralNetConfiguration conf,
                         StepFunction stepFunction,
                         Collection<IterationListener> iterationListeners,
                         Model model)

ConjugateGradient

public ConjugateGradient(NeuralNetConfiguration conf,
                         StepFunction stepFunction,
                         Collection<IterationListener> iterationListeners,
                         Collection<TerminationCondition> terminationConditions,
                         Model model)

Method Detail

preProcessLine

public void preProcessLine(org.nd4j.linalg.api.ndarray.INDArray line)

Description copied from class: BaseOptimizer

Pre process the line (scaling and the like)

Specified by:

preProcessLine in interface ConvexOptimizer

Overrides:

preProcessLine in class BaseOptimizer

Parameters:

line - the line to pre process

postStep

public void postStep()

Description copied from class: BaseOptimizer

Post step (conjugate gradient among other methods needs this)

Specified by:

postStep in interface ConvexOptimizer

Overrides:

postStep in class BaseOptimizer

setupSearchState

public void setupSearchState(Pair<Gradient,Double> pair)

Description copied from class: BaseOptimizer

Setup the initial search state

Specified by:

setupSearchState in interface ConvexOptimizer

Overrides:

setupSearchState in class BaseOptimizer

Parameters:

pair - the gradient and score