de.jstacs.sampling

Interface GibbsSamplingModel

All Superinterfaces:

SamplingComponent

All Known Implementing Classes:

FSDAGModelForGibbsSampling

public interface GibbsSamplingModel
extends SamplingComponent

This is the interface that any AbstractTrainableStatisticalModel has to implement if it should be used in a sampling.

During the sampling the method drawParameters(DataSet, double[] ) is used for drawing the parameters from the posterior.

Author:

Berit Haldemann, Jens Keilwagen

See Also:

AbstractTrainableStatisticalModel, AbstractMixtureTrainSM, AbstractMixtureTrainSM.Algorithm.GIBBS_SAMPLING

Method Summary

Modifier and Type	Method and Description
void	drawParameters(DataSet data, double[] weights) This method draws the parameters of the model from the a posteriori density.

Methods inherited from interface de.jstacs.sampling.SamplingComponent

acceptParameters, extendSampling, initForSampling, isInSamplingMode, parseNextParameterSet, parseParameterSet, samplingStopped

Method Detail

drawParameters

void drawParameters(DataSet data,
                    double[] weights)
             throws Exception

This method draws the parameters of the model from the a posteriori density. It is recommended to write the parameters to a specific file using SamplingComponent.acceptParameters() so that they can later be parsed using the methods of the interface.

Before using this method the method SamplingComponent.initForSampling(int) should be called.

Parameters:

data - a data set

weights - the (non-negative) weights for each sequence of the data set

Throws:

Exception - if there is a problem with drawing the parameters, the model is not initialized, ...

See Also:

SamplingComponent.initForSampling(int), SamplingComponent.acceptParameters()