de.jstacs.classifier.scoringFunctionBased
Class OptimizableFunction

java.lang.Object
  de.jstacs.algorithms.optimization.DifferentiableFunction
      de.jstacs.classifier.scoringFunctionBased.OptimizableFunction

All Implemented Interfaces:

Function

Direct Known Subclasses:

AbstractOptimizableFunction

public abstract class OptimizableFunction
extends DifferentiableFunction

This is the main function for the ScoreClassifier.

Author:

Jens Keilwagen

See Also:

Optimizer

Nested Class Summary

static class

OptimizableFunction.KindOfParameter This enum defines the kinds of parameters that can be returned by the method getParameters(KindOfParameter).

Constructor Summary

OptimizableFunction()

Method Summary

abstract void	addTermToClassParameter(int classIndex, double term) This method adds the term to the class parameter.
abstract double[]	getClassParams(double[] params) Returns from the complete vector of parameters those that are for the classes.
abstract int	getNumberOfStarts() Returns the number of starts that should be done for a good optimum.
abstract double[]	getParameters(OptimizableFunction.KindOfParameter kind) Returns some parameters that can be used for instance as start parameters.
abstract void	reset(ScoringFunction[] score) Resets the ScoringFunctions and all pre-computed values.
abstract void	setParams(double[] current) Sets the current values as parameters.

Methods inherited from class de.jstacs.algorithms.optimization.DifferentiableFunction

evaluateGradientOfFunction, findOneDimensionalMin

Methods inherited from class java.lang.Object

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

Methods inherited from interface de.jstacs.algorithms.optimization.Function

evaluateFunction, getDimensionOfScope

Constructor Detail

OptimizableFunction

public OptimizableFunction()

Method Detail

getNumberOfStarts

public abstract int getNumberOfStarts()

Returns the number of starts that should be done for a good optimum. For concave or convex functions this number should be 1.

Returns:

the number of starts the Optimizer should be started to get a good optimum

See Also:

Optimizer

getClassParams

public abstract double[] getClassParams(double[] params)

Returns from the complete vector of parameters those that are for the classes.

Parameters:

params - the current parameters

Returns:

the parameters for the classes

getParameters

public abstract double[] getParameters(OptimizableFunction.KindOfParameter kind)
                                throws Exception

Returns some parameters that can be used for instance as start parameters.

Parameters:

kind - the kind of the class parameters that will be returned

Returns:

some start parameters

Throws:

Exception - if something went wrong

setParams

public abstract void setParams(double[] current)
                        throws DimensionException

Sets the current values as parameters.

Parameters:

current - the current values

Throws:

DimensionException - if the dimension of the current values does not match with the internal parameters

reset

public abstract void reset(ScoringFunction[] score)
                    throws Exception

Resets the ScoringFunctions and all pre-computed values.

Parameters:

score - the array of ScoringFunctions

Throws:

Exception - if something went wrong

addTermToClassParameter

public abstract void addTermToClassParameter(int classIndex,
                                             double term)

This method adds the term to the class parameter.

Parameters:

classIndex - the index of the class

term - the term to be added