de.jstacs.models.discrete.inhomogeneous.shared
Class SharedStructureMixture

java.lang.Object
  de.jstacs.models.AbstractModel
      de.jstacs.models.mixture.AbstractMixtureModel
          de.jstacs.models.mixture.MixtureModel
              de.jstacs.models.discrete.inhomogeneous.shared.SharedStructureMixture

All Implemented Interfaces:

Model, Storable, Cloneable

public class SharedStructureMixture
extends MixtureModel

This class handles a mixture of models with the same structure that are learned via EM. One well known example are mixture of trees.

Author:

Jens Keilwagen

Nested Class Summary

Nested classes/interfaces inherited from class de.jstacs.models.mixture.AbstractMixtureModel

AbstractMixtureModel.Algorithm, AbstractMixtureModel.Parameterization

Field Summary

Fields inherited from class de.jstacs.models.mixture.AbstractMixtureModel

algorithm, algorithmHasBeenRun, alternativeModel, burnInTest, componentHyperParams, compProb, counter, dimension, estimateComponentProbs, file, filereader, filewriter, initialIteration, logWeights, model, optimizeModel, sample, samplingIndex, sostream, starts, stationaryIteration, weights

Fields inherited from class de.jstacs.models.AbstractModel

alphabets, length

Constructor Summary

protected	SharedStructureMixture(FSDAGModel[] m, StructureLearner.ModelType model, byte order, int starts, boolean estimateComponentProbs, double[] weights, double alpha, double epsilon) This constructor is used from the other main constructors.
	SharedStructureMixture(FSDAGModel[] m, StructureLearner.ModelType model, byte order, int starts, double[] weights, double alpha, double epsilon) This main constructor creates an instance with fixed component weights.
	SharedStructureMixture(FSDAGModel[] m, StructureLearner.ModelType model, byte order, int starts, double alpha, double epsilon) This main constructor creates an instance which estimates the component probabilities.
	SharedStructureMixture(StringBuffer xml) The constructor for the Storable interface.

Method Summary

SharedStructureMixture	clone() Follows the conventions of Object's clone-method.
protected void	fromXML(StringBuffer representation) This method should only be used by the constructor that works on StringBuffer.
String	getInstanceName() Should return a short instance name such as iMM(0), BN(2), ...
protected void	getNewParameters(int iteration, double[][] seqWeights, double[] w) This method trains the internal models on the internal sample and the given weights.
String	getStructure()
StringBuffer	toXML() This method returns an XML-representation of an instance of the implementing class.

Methods inherited from class de.jstacs.models.mixture.MixtureModel

doFirstIteration, doFirstIteration, emitSampleUsingCurrentParameterSet, getLogProbUsingCurrentParameterSetFor, getNewWeights, setTrainData, toString

Methods inherited from class de.jstacs.models.mixture.AbstractMixtureModel

algorithmHasBeenRun, checkLength, checkModelsForGibbsSampling, continueIterations, continueIterations, createSeqWeightsArray, doFirstIteration, doFirstIteration, draw, emitSample, extendSampling, extractFurtherInformation, finalize, getCharacteristics, getFurtherInformation, getIndexOfMaximalComponentFor, getLogPriorTerm, getLogPriorTermForComponentProbs, getLogProbFor, getLogProbFor, getLogProbFor, getModel, getModels, getMRG, getMRGParams, getNameOfAlgorithm, getNewComponentProbs, getNewParametersForModel, getNumberOfComponents, getNumericalCharacteristics, getProbFor, getScoreForBestRun, getWeights, initModelForSampling, initWithPrior, isInSamplingMode, isTrained, iterate, iterate, max, modifyWeights, parseNextParameterSet, parseParameterSet, samplingStopped, set, setAlpha, setOutputStream, setThreshold, setWeights, swap, train

Methods inherited from class de.jstacs.models.AbstractModel

getAlphabetContainer, getLength, getLogProbFor, getLogProbFor, getLogProbFor, getMaximalMarkovOrder, getPriorTerm, getProbFor, getProbFor, setNewAlphabetContainerInstance, train

Methods inherited from class java.lang.Object

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

Constructor Detail

SharedStructureMixture

public SharedStructureMixture(FSDAGModel[] m,
                              StructureLearner.ModelType model,
                              byte order,
                              int starts,
                              double alpha,
                              double epsilon)
                       throws IllegalArgumentException,
                              WrongAlphabetException,
                              CloneNotSupportedException

This main constructor creates an instance which estimates the component probabilities.

Parameters:

m - the single models building the Mixture

model - the model type

order - the model order

starts - the number of times the algorithm will be started in the train method, at least 1

alpha - the positive parameter for the Dirichlet which is used when you invoke train to initialize the gammas. It is recommended to use alpha = 1 (uniform distribution on a simplex).

epsilon - the non-negative threshold for stopping the EM-algorithm

Throws:

IllegalArgumentException - if

• the models are not able to score the sequences of same length
• dimension < 1
• weights != null && weights.length != dimension
• weights != null and it exists an i where weights[i] < 0
• starts < 1
• componentHyperParams are not correct

WrongAlphabetException - if not all models work on the same alphabet

CloneNotSupportedException - if the models can not be cloned

SharedStructureMixture

public SharedStructureMixture(FSDAGModel[] m,
                              StructureLearner.ModelType model,
                              byte order,
                              int starts,
                              double[] weights,
                              double alpha,
                              double epsilon)
                       throws IllegalArgumentException,
                              WrongAlphabetException,
                              CloneNotSupportedException

This main constructor creates an instance with fixed component weights.

Parameters:

m - the single models building the Mixture

model - the model type

order - the model order

starts - the number of times the algorithm will be started in the train method, at least 1

weights - null or the weights for the components (than weights.length == models.length)

alpha - the positive parameter for the Dirichlet which is used when you invoke train to initialize the gammas. It is recommended to use alpha = 1 (uniform distribution on a simplex).

epsilon - the non-negative threshold for stopping the EM-algorithm

Throws:

IllegalArgumentException - if

• the models are not able to score the sequences of same length
• dimension < 1
• weights != null && weights.length != dimension
• weights != null and it exists an i where weights[i] < 0
• starts < 1
• componentHyperParams are not correct

WrongAlphabetException - if not all models work on the same alphabet

CloneNotSupportedException - if the models can not be cloned

SharedStructureMixture

protected SharedStructureMixture(FSDAGModel[] m,
                                 StructureLearner.ModelType model,
                                 byte order,
                                 int starts,
                                 boolean estimateComponentProbs,
                                 double[] weights,
                                 double alpha,
                                 double epsilon)
                          throws IllegalArgumentException,
                                 WrongAlphabetException,
                                 CloneNotSupportedException

This constructor is used from the other main constructors.

Parameters:

m - the single models building the Mixture

model - the model type

order - the model order

starts - the number of times the algorithm will be started in the train method, at least 1

estimateComponentProbs - the switch for estimating the component probabilities in the algorithm or to hold them fixed; if the component parameters are fixed, the values of weights will be used, otherwise the componentHyperParams will be incorporated in the adjustment

weights - null or the weights for the components (than weights.length == models.length)

alpha - the positive parameter for the Dirichlet which is used when you invoke train to initialize the gammas. It is recommended to use alpha = 1 (uniform distribution on a simplex).

epsilon - the non-negative threshold for stopping the EM-algorithm

Throws:

IllegalArgumentException - if

• the models are not able to score the sequences of same length
• dimension < 1
• weights != null && weights.length != dimension
• weights != null and it exists an i where weights[i] < 0
• starts < 1
• componentHyperParams are not correct

WrongAlphabetException - if not all models work on the same alphabet

CloneNotSupportedException - if the models can not be cloned

SharedStructureMixture

public SharedStructureMixture(StringBuffer xml)
                       throws NonParsableException

The constructor for the Storable interface.

Parameters:

xml - the StringBuffer

Throws:

NonParsableException - if the StringBuffer is not parsable

Method Detail

clone

public SharedStructureMixture clone()
                             throws CloneNotSupportedException

Description copied from class: AbstractModel

Follows the conventions of Object's clone-method.

Specified by:

clone in interface Model

Overrides:

clone in class AbstractMixtureModel

Returns:

an object, that is a copy of the current AbstractModel (the member-AlphabetContainer isn't deeply cloned since it is assumed to be immutable). The type of the returned object is defined by the class X directly inherited from AbstractModel. Hence X's clone-method should work as:
1. Object o = (X)super.clone(); 2. all additional member variables of o defined by X that are not of simple data-types like int, double, ... , have to be deeply copied 3. return o

Throws:

CloneNotSupportedException

getStructure

public String getStructure()
                    throws NotTrainedException

Returns:

a string representation of the structure of the used models

Throws:

NotTrainedException - if the classifier is not trained yet

getInstanceName

public String getInstanceName()

Description copied from interface: Model

Should return a short instance name such as iMM(0), BN(2), ...

Specified by:

getInstanceName in interface Model

Overrides:

getInstanceName in class AbstractMixtureModel

Returns:

a short instance name

toXML

public StringBuffer toXML()

Description copied from interface: Storable

This method returns an XML-representation of an instance of the implementing class.

Specified by:

toXML in interface Storable

Overrides:

toXML in class AbstractMixtureModel

Returns:

the XML-representation

fromXML

protected void fromXML(StringBuffer representation)
                throws NonParsableException

Description copied from class: AbstractModel

This method should only be used by the constructor that works on StringBuffer. It is the counter part of toXML().

Overrides:

fromXML in class AbstractMixtureModel

Parameters:

representation - the representation

Throws:

NonParsableException - if the StringBuffer is not parsable or the representation is conflicting

See Also:

AbstractModel.AbstractModel(StringBuffer)

getNewParameters

protected void getNewParameters(int iteration,
                                double[][] seqWeights,
                                double[] w)
                         throws Exception

Description copied from class: AbstractMixtureModel

This method trains the internal models on the internal sample and the given weights.

Overrides:

getNewParameters in class AbstractMixtureModel

Parameters:

iteration - the number this method has been invoked

seqWeights - the weights for each model and sequence

w - the weights for the components

Throws:

Exception - if the training of the internal models went wrong