weka.classifiers.bayes.blr

Class Prior

java.lang.Object

weka.classifiers.bayes.blr.Prior

All Implemented Interfaces:

java.io.Serializable, RevisionHandler

Direct Known Subclasses:

GaussianPriorImpl, LaplacePriorImpl

public abstract class Prior
extends java.lang.Object
implements java.io.Serializable, RevisionHandler

This is an interface to plug various priors into the Bayesian Logistic Regression Model.

Version:

$Revision: 1.2 $

Author:

Navendu Garg (gargnav@iit.edu)

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
Prior()

Method Summary

Modifier and Type	Method and Description
void	computelogLikelihood(double[] betas, Instances instances) Function computes the log-likelihood value: -sum{1 to n}{ln(1+exp(-Beta*x(i)*y(i))}
void	computePenalty(double[] betas, double[] hyperparameters) Skeleton function to compute penalty terms.
double	getLoglikelihood()
double	getLogPosterior()
double	getPenalty()
double	update(int j, Instances instances, double beta, double hyperparameter, double[] r, double deltaV) Interface for the update functions for different types of priors.

Methods inherited from class java.lang.Object

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

Methods inherited from interface weka.core.RevisionHandler

getRevision

Constructor Detail

Prior

public Prior()

Method Detail

update

public double update(int j,
                     Instances instances,
                     double beta,
                     double hyperparameter,
                     double[] r,
                     double deltaV)

Interface for the update functions for different types of priors.

computelogLikelihood

public void computelogLikelihood(double[] betas,
                                 Instances instances)

Function computes the log-likelihood value: -sum{1 to n}{ln(1+exp(-Beta*x(i)*y(i))}

Parameters:

betas -

instances -

computePenalty

public void computePenalty(double[] betas,
                           double[] hyperparameters)

Skeleton function to compute penalty terms.

Parameters:

betas -

hyperparameters -

getLoglikelihood

public double getLoglikelihood()

Returns:

log-likelihood value.

getLogPosterior

public double getLogPosterior()

Returns:

regularized log posterior value.

getPenalty

public double getPenalty()

Returns:

penalty term.