edu.stanford.nlp.optimization

Class SGDWithAdaGradAndFOBOS<T extends Function>

java.lang.Object

edu.stanford.nlp.optimization.SGDWithAdaGradAndFOBOS<T>

All Implemented Interfaces:

HasEvaluators, Minimizer<T>

public class SGDWithAdaGradAndFOBOS<T extends Function>
extends Object
implements Minimizer<T>, HasEvaluators

Stochastic Gradient Descent With AdaGrad and FOBOS in batch mode. Similar to SGDMinimizer, regularization is done in the minimizer, not in the objective function. This version only does batch optimization. For online variant, see SparseAdaGradMinimizer.java

Author:

Mengqiu Wang

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	SGDWithAdaGradAndFOBOS.Prior

Field Summary

Fields

Modifier and Type	Field and Description
protected double	alpha
protected int	bSize
protected Random	gen
protected double	initRate
protected double	lambda
protected long	maxTime
protected int	numPasses
protected boolean	quiet
protected double[]	x

Constructor Summary

Constructors

Constructor and Description
SGDWithAdaGradAndFOBOS(double initRate, double lambda, int numPasses)
SGDWithAdaGradAndFOBOS(double initRate, double lambda, int numPasses, int batchSize)
SGDWithAdaGradAndFOBOS(double initRate, double lambda, int numPasses, int batchSize, String priorType, double alpha)

Method Summary

Methods

Modifier and Type	Method and Description
protected String	getName()
double[]	minimize(Function function, double functionTolerance, double[] initial) Attempts to find an unconstrained minimum of the objective function starting at initial, within functionTolerance.
double[]	minimize(Function f, double functionTolerance, double[] initial, int maxIterations)
protected void	say(String s)
protected void	sayln(String s)
void	setEvaluators(int iters, Evaluator[] evaluators)
void	setTerminateOnEvalImprovementNumOfEpoch(int terminateOnEvalImprovementNumOfEpoch)
void	shutUp()
void	suppressTestPrompt(boolean suppressTestPrompt)
void	terminateOnEvalImprovement(boolean toTerminate)
boolean	toContinue(double[] x, double currEval)

Methods inherited from class java.lang.Object

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

Field Detail

x

protected double[] x

initRate

protected double initRate

lambda

protected double lambda

alpha

protected double alpha

quiet

protected boolean quiet

numPasses

protected final int numPasses

bSize

protected int bSize

gen

protected Random gen

maxTime

protected long maxTime

Constructor Detail

SGDWithAdaGradAndFOBOS

public SGDWithAdaGradAndFOBOS(double initRate,
                      double lambda,
                      int numPasses)

SGDWithAdaGradAndFOBOS

public SGDWithAdaGradAndFOBOS(double initRate,
                      double lambda,
                      int numPasses,
                      int batchSize)

SGDWithAdaGradAndFOBOS

public SGDWithAdaGradAndFOBOS(double initRate,
                      double lambda,
                      int numPasses,
                      int batchSize,
                      String priorType,
                      double alpha)

Method Detail

terminateOnEvalImprovement

public void terminateOnEvalImprovement(boolean toTerminate)

suppressTestPrompt

public void suppressTestPrompt(boolean suppressTestPrompt)

setTerminateOnEvalImprovementNumOfEpoch

public void setTerminateOnEvalImprovementNumOfEpoch(int terminateOnEvalImprovementNumOfEpoch)

toContinue

public boolean toContinue(double[] x,
                 double currEval)

shutUp

public void shutUp()

getName

protected String getName()

setEvaluators

public void setEvaluators(int iters,
                 Evaluator[] evaluators)

Specified by:

setEvaluators in interface HasEvaluators

minimize

public double[] minimize(Function function,
                double functionTolerance,
                double[] initial)

Description copied from interface: Minimizer

Attempts to find an unconstrained minimum of the objective function starting at initial, within functionTolerance.

Specified by:

minimize in interface Minimizer<T extends Function>

Parameters:

function - the objective function

functionTolerance - a double value

initial - a initial feasible point

Returns:

Unconstrained minimum of function

minimize

public double[] minimize(Function f,
                double functionTolerance,
                double[] initial,
                int maxIterations)

Specified by:

minimize in interface Minimizer<T extends Function>

sayln

protected void sayln(String s)

say

protected void say(String s)