de.jstacs.classifiers.differentiableSequenceScoreBased.logPrior

Class LogPrior

java.lang.Object

de.jstacs.algorithms.optimization.DifferentiableFunction

de.jstacs.classifiers.differentiableSequenceScoreBased.logPrior.LogPrior

All Implemented Interfaces:

Function, Storable

Direct Known Subclasses:

CompositeLogPrior, DoesNothingLogPrior, SeparateLogPrior, SimpleGaussianSumLogPrior

public abstract class LogPrior
extends DifferentiableFunction
implements Storable

The abstract class for any log-prior used e.g. for maximum supervised posterior optimization.

Author:

Jens Keilwagen

Field Summary

Fields

Modifier and Type	Field and Description
static int	UNKNOWN Sometimes the method Function.getDimensionOfScope() can not return a value.

Constructor Summary

Constructors

Constructor and Description
LogPrior()

Method Summary

Modifier and Type	Method and Description
abstract void	addGradientFor(double[] params, double[] vector) Adds the gradient of the log-prior using the current parameters to a given vector.
double[]	evaluateGradientOfFunction(double[] params) Evaluates the gradient of a function at a certain vector (in mathematical sense) x, i.e., .
abstract String	getInstanceName() Returns a short instance name.
abstract LogPrior	getNewInstance() This method returns an empty new instance of the current prior.
void	set(boolean freeParameters, DifferentiableSequenceScore... funs) Resets all pre-computed values to their initial values using the DifferentiableSequenceScores funs.
abstract StringBuffer	toXML() Encodes the prior as an XML representation.

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

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

Field Detail

UNKNOWN

public static final int UNKNOWN

Sometimes the method Function.getDimensionOfScope() can not return a value. In these cases it is recommended to use this constant.

See Also:

Constant Field Values

Constructor Detail

LogPrior

public LogPrior()

Method Detail

addGradientFor

public abstract void addGradientFor(double[] params,
                                    double[] vector)
                             throws EvaluationException

Adds the gradient of the log-prior using the current parameters to a given vector.

Parameters:

params - the parameters

vector - the vector

Throws:

EvaluationException - if the gradient could not be evaluated

evaluateGradientOfFunction

public final double[] evaluateGradientOfFunction(double[] params)
                                          throws EvaluationException

Description copied from class: DifferentiableFunction

Evaluates the gradient of a function at a certain vector (in mathematical sense) x, i.e., .

Specified by:

evaluateGradientOfFunction in class DifferentiableFunction

Parameters:

params - the current vector

Returns:

the evaluation of the gradient of a function, has dimension Function.getDimensionOfScope()

Throws:

EvaluationException - if there was something wrong during the evaluation of the gradient

See Also:

Function.getDimensionOfScope()

getNewInstance

public abstract LogPrior getNewInstance()
                                 throws CloneNotSupportedException

This method returns an empty new instance of the current prior. The method works similar to clone, but does not clone DifferentiableSequenceScore s that may be inside the instance. The DifferentiableSequenceScores must be set by an invocation of the method set(boolean, DifferentiableSequenceScore...).

Returns:

a new empty instance of the prior

Throws:

CloneNotSupportedException - if something went wrong while cloning

See Also:

set(boolean, DifferentiableSequenceScore...)

set

public void set(boolean freeParameters,
                DifferentiableSequenceScore... funs)
         throws Exception

Resets all pre-computed values to their initial values using the DifferentiableSequenceScores funs.

Parameters:

freeParameters - the switch for using only the free parameters or all parameters in a DifferentiableSequenceScore

funs - the DifferentiableSequenceScores for the prior

Throws:

Exception - if the DifferentiableSequenceScores could not be set

toXML

public abstract StringBuffer toXML()

Encodes the prior as an XML representation. It does not encode all information, since the method set(boolean, DifferentiableSequenceScore...) has to be invoked after decoding.

Specified by:

toXML in interface Storable

Returns:

the XML representation;

getInstanceName

public abstract String getInstanceName()

Returns a short instance name.

Returns:

a short instance name