de.jstacs.classifier.scoringFunctionBased
Class AbstractOptimizableFunction

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

All Implemented Interfaces:

Function

Direct Known Subclasses:

NormConditionalLogLikelihood

public abstract class AbstractOptimizableFunction
extends OptimizableFunction

This class extends OptimizableFunction and implements some common methods.

Author:

Jens Keilwagen

Nested Class Summary

Nested classes/interfaces inherited from class de.jstacs.classifier.scoringFunctionBased.OptimizableFunction

OptimizableFunction.KindOfParameter

Field Summary

protected int	cl The number of different classes.
protected double[]	clazz The class parameters.
protected Sample[]	data The data that is used to evaluate this function.
protected boolean	freeParams Whether only the free parameters or all should be used.
protected double[]	logClazz The logarithm of the class parameters.
protected boolean	norm Whether a normalization should be done or not.
protected int[]	shortcut These shortcuts indicate the beginning of a new part in the parameter vector.
protected double[]	sum The sums of the weighted data per class and additional the total weight sum.
protected double[][]	weights The weights for the data.

Constructor Summary

protected

AbstractOptimizableFunction(Sample[] data, double[][] weights, boolean norm, boolean freeParams) The constructor creates an instance using the given weighted data.

Method Summary

void	addTermToClassParameter(int classIndex, double term) This method adds the term to the class parameter.
double[]	getClassParams(double[] params) Returns from the complete vector of parameters those that are for the classes.
int	getDimensionOfScope() Returns the dimension of the scope of the function.
protected int	getNumberOfStarts(ScoringFunction[] score) Returns the number of recommended starts.
double[]	getParameters(OptimizableFunction.KindOfParameter kind) Returns some parameters that can be used for instance as start parameters.
abstract void	getParameters(OptimizableFunction.KindOfParameter kind, double[] erg) This method enables the user to get the parameters without creating a new array.
void	setParams(double[] params) Checks the dimension and sets the class parameters.

Methods inherited from class de.jstacs.classifier.scoringFunctionBased.OptimizableFunction

getNumberOfStarts, reset

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

Field Detail

shortcut

protected int[] shortcut

These shortcuts indicate the beginning of a new part in the parameter vector.

data

protected Sample[] data

The data that is used to evaluate this function.

weights

protected double[][] weights

The weights for the data.

See Also:

data

clazz

protected double[] clazz

The class parameters.

logClazz

protected double[] logClazz

The logarithm of the class parameters.

See Also:

clazz

sum

protected double[] sum

The sums of the weighted data per class and additional the total weight sum.

See Also:

data, weights

cl

protected int cl

The number of different classes.

norm

protected boolean norm

Whether a normalization should be done or not.

freeParams

protected boolean freeParams

Whether only the free parameters or all should be used.

Constructor Detail

AbstractOptimizableFunction

protected AbstractOptimizableFunction(Sample[] data,
                                      double[][] weights,
                                      boolean norm,
                                      boolean freeParams)
                               throws IllegalArgumentException,
                                      WrongAlphabetException

The constructor creates an instance using the given weighted data.

Parameters:

data - the data

weights - the weights

norm - the switch for using the normalization (division by the number of sequences)

freeParams - the switch for using only the free parameters

Throws:

IllegalArgumentException - if the number of classes is not correct

WrongAlphabetException - if different alphabets are used

Method Detail

getDimensionOfScope

public final int getDimensionOfScope()

Description copied from interface: Function

Returns the dimension of the scope of the function.

Returns:

dimension of the scope; n, with f: R^n -> R

getParameters

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

This method enables the user to get the parameters without creating a new array.

Parameters:

kind - the kind of the class parameters to be returned in erg

erg - the array for the start parameters

Throws:

Exception - if the array is null or does not have the correct length

See Also:

OptimizableFunction.getParameters(KindOfParameter)

getParameters

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

Description copied from class: OptimizableFunction

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

Specified by:

getParameters in class OptimizableFunction

Parameters:

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

Returns:

some start parameters

Throws:

Exception - if something went wrong

setParams

public void setParams(double[] params)
               throws DimensionException

Checks the dimension and sets the class parameters.

Specified by:

setParams in class OptimizableFunction

Parameters:

params - the current values

Throws:

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

getClassParams

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

Description copied from class: OptimizableFunction

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

Specified by:

getClassParams in class OptimizableFunction

Parameters:

params - the current parameters

Returns:

the parameters for the classes

getNumberOfStarts

protected final int getNumberOfStarts(ScoringFunction[] score)

Returns the number of recommended starts.

Parameters:

score - the underlying scoring functions

Returns:

the number of recommended starts

See Also:

OptimizableFunction.getNumberOfStarts()

addTermToClassParameter

public final void addTermToClassParameter(int classIndex,
                                          double term)

Description copied from class: OptimizableFunction

This method adds the term to the class parameter.

Specified by:

addTermToClassParameter in class OptimizableFunction

Parameters:

classIndex - the index of the class

term - the term to be added