weka.classifiers.bayes.blr
Class GaussianPriorImpl

java.lang.Object
  extended by weka.classifiers.bayes.blr.Prior
      extended by weka.classifiers.bayes.blr.GaussianPriorImpl
All Implemented Interfaces:
java.io.Serializable, RevisionHandler

public class GaussianPriorImpl
extends Prior

Implementation of the Gaussian Prior update function based on CLG Algorithm with a certain Trust Region Update. The values are updated in the BayesianLogisticRegressionV variables used by the algorithm.

Version:
$Revision: 1.2 $
Author:
Navendu Garg(gargnav@iit.edu)
See Also:
Serialized Form

Constructor Summary
GaussianPriorImpl()
           
 
Method Summary
 void computeLoglikelihood(double[] betas, Instances instances)
          This method calls the log-likelihood implemented in the Prior abstract class.
 void computePenalty(double[] betas, double[] hyperparameters)
          This function computes the penalty term specific to Gaussian distribution.
 java.lang.String getRevision()
          Returns the revision string.
 double update(int j, Instances instances, double beta, double hyperparameter, double[] r, double deltaV)
          Update function specific to Laplace Prior.
 
Methods inherited from class weka.classifiers.bayes.blr.Prior
computelogLikelihood, getLoglikelihood, getLogPosterior, getPenalty
 
Methods inherited from class java.lang.Object
equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Constructor Detail

GaussianPriorImpl

public GaussianPriorImpl()
Method Detail

update

public double update(int j,
                     Instances instances,
                     double beta,
                     double hyperparameter,
                     double[] r,
                     double deltaV)
Update function specific to Laplace Prior.

Overrides:
update in class Prior

computeLoglikelihood

public void computeLoglikelihood(double[] betas,
                                 Instances instances)
This method calls the log-likelihood implemented in the Prior abstract class.

Parameters:
betas -
instances -

computePenalty

public void computePenalty(double[] betas,
                           double[] hyperparameters)
This function computes the penalty term specific to Gaussian distribution.

Overrides:
computePenalty in class Prior
Parameters:
betas -
hyperparameters -

getRevision

public java.lang.String getRevision()
Returns the revision string.

Returns:
the revision