Package de.jstacs.classifier.scoringFunctionBased.sampling

Provides the classes for AbstractScoreBasedClassifiers that are based on SamplingScoringFunctions and that sample parameters using the Metropolis-Hastings algorithm.

See:
Description

Class Summary

SamplingGenDisMixClassifier	A classifier that samples its parameters from a LogGenDisMixFunction using the Metropolis-Hastings algorithm.
SamplingGenDisMixClassifierParameterSet	ParameterSet to instantiate a SamplingGenDisMixClassifier.
SamplingScoreBasedClassifier	A classifier that samples the parameters of SamplingScoringFunctions by the Metropolis-Hastings algorithm.
SamplingScoreBasedClassifierParameterSet	ParameterSet to instantiate a SamplingScoreBasedClassifier.

Enum Summary

SamplingScoreBasedClassifier.SamplingScheme

Sampling scheme for sampling the parameters of the scoring functions.

Package de.jstacs.classifier.scoringFunctionBased.sampling Description

Provides the classes for AbstractScoreBasedClassifiers that are based on SamplingScoringFunctions and that sample parameters using the Metropolis-Hastings algorithm. The abstract base class of all such classifiers is the SamplingScoreBasedClassifier, while the SamplingGenDisMixClassifier provides an implementation for the unified generative-discriminative learning principle.

Since:

Jstacs 1.5

See Also:

SamplingScoreBasedClassifier, SamplingGenDisMixClassifier, SamplingScoringFunction