com.mockturtlesolutions.snifflib.invprobs

Class DistributedStatisticalModel

java.lang.Object

com.mockturtlesolutions.snifflib.functions.AbstractFunction

com.mockturtlesolutions.snifflib.invprobs.StatisticalModel

com.mockturtlesolutions.snifflib.invprobs.DistributedStatisticalModel

All Implemented Interfaces:

Nameable, FigureReportable, Reportable, DeclaredParameters, DeclaredVariables, Likelihood, NamedParameters, ObjectSampler, ObjectSamplerClient, OptimizableScalar

public abstract class DistributedStatisticalModel
extends StatisticalModel

Provides a transparent extension of StatisticalModel for cases where parameters and/or data may be distributed across one or more other classes (e.g. storage classes, datasets, JTable etc). Note that a MethodStorage can be used to store parameter mappings. It is desirable that the map only relates to specific mapped parameters because for efficiency sake we do not want to write methods that have to parse parameter names every time a parameter is desired to be gotten or set.

Field Summary

Fields

Modifier and Type	Field and Description
protected java.util.HashMap	methodMap

Fields inherited from class com.mockturtlesolutions.snifflib.invprobs.StatisticalModel

adapter, FigureReporters, invertParameterDeclarations, LIKELIHOOD, NEG_LOG_LIKELIHOOD, paramListeners, Params, PosteriorParams, PREDICTION, priorParamListeners, PriorParams, SUM_NEG_LOG_LIKELIHOOD, usingDataSet

Fields inherited from class com.mockturtlesolutions.snifflib.functions.AbstractFunction

domainDimension, rangeDimension

Constructor Summary

Constructors

Constructor and Description
DistributedStatisticalModel()

Method Summary

Methods

Modifier and Type	Method and Description
static java.lang.String	convertDblToString(DblMatrix in)
static DblMatrix	convertStringToDbl(java.lang.String in)
protected abstract void	declareDistributedParameter(java.lang.String p)
protected void	declareDistributedParameter(java.lang.String p, DblMatrix v)
protected boolean	declareIfNotDistributedParameter(java.lang.String p, DblMatrix v)
void	declareParameter(java.lang.String p)
void	declareParameter(java.lang.String p, DblMatrix v) Convenience method that declares a parameter and sets a initial (default) value for that parameter.
void	declareParameterSuper(java.lang.String p)
void	declareParameterSuper(java.lang.String p, DblMatrix v)
int	distributedNumParams()
java.lang.String[]	distributedParameterSet()
java.util.Set	getDeclaredDistributedParameters()
java.util.Set	getDeclaredParameters()
DblMatrix	getDistributedParam(java.lang.String paramname)
DblParamSet	getDistributedParams()
DblMatrix	getParam(java.lang.String name)
DblParamSet	getParams()
DblMatrix	getParamSuper(java.lang.String name) Provides access to the StatisticalModel super class method getParam.
boolean	hasDistributedParameter(java.lang.String s)
boolean	hasParameter(java.lang.String s)
int	numParams() Return the number of parameters.
java.lang.String[]	parameterSet()
void	replaceDistributedParams(DblParamSet X)
void	replaceParams(DblParamSet X) Replace parameters with those in X.
protected void	revokeDistributedParameter(java.lang.String p)
void	revokeParameter(java.lang.String p)
void	revokeParameterSuper(java.lang.String p)
void	setDistributedParam(java.lang.String paramname, DblMatrix value)
void	setParam(java.lang.String name, DblMatrix value)
void	setParamSuper(java.lang.String name, DblMatrix value) Provides access to the StatisticalModel super class method setParam.

Methods inherited from class com.mockturtlesolutions.snifflib.invprobs.StatisticalModel

about, addFigureReporter, addObjectToSample, addParameterChangeListener, declareIfNotParameter, declareIfNotPriorParameter, declarePriorParameter, declarePriorParameter, declareVariable, fireParameterAdded, fireParameterRemoved, fireParameterValueChanged, fireParameterValuesChanged, firePriorParameterAdded, firePriorParameterRemoved, firePriorParameterValueChanged, firePriorParameterValuesChanged, getDataAugment, getDataSet, getDataSetAdapter, getDeclaredPriorParameters, getDeclaredVariables, getFigureReporter, getFigureReporters, getIgnoreNaN, getMSE, getName, getNewObjectSample, getNewObjectSamples, getObjectSamplingOffered, getObjectsToSample, getPosteriorParam, getPosteriorParams, getPredictionAt, getPriorParam, getPriorParams, getValueAt, getValueAt, getValueAt, getValueToMinimize, getX, getY, hasPriorParameter, invertParameterDeclarations, isSamplingObject, isVectorBased, likelihood, mapName, negLogLikelihood, negLogPrior, offerObjectSamplingFor, offersObjectSamplingFor, putNewObjectSample, removeAllParameterChangeListener, removeFigureReporter, removeObjectToSample, removeParameterChangeListener, replacePosteriorParams, replacePriorParams, residuals, revokeObjectSamplingFor, revokePriorParameter, revokeVariable, samplePosterior, samplePosterior, setDataAugment, setDataSet, setIgnoreNaN, setName, setObjectsToSample, setParams, setPosteriorParam, setPosteriorParams, setPriorParam, setPriorParams, setValueToGet, setX, setY, sumNegLogLikelihood, unMapName, variableSet

Methods inherited from class com.mockturtlesolutions.snifflib.functions.AbstractFunction

domainDimension, plotOver, plotOver, rangeDimension

Methods inherited from class java.lang.Object

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

Field Detail

methodMap

protected java.util.HashMap methodMap

Constructor Detail

DistributedStatisticalModel

public DistributedStatisticalModel()

Method Detail

numParams

public int numParams()

Description copied from class: StatisticalModel

Return the number of parameters.

Overrides:

numParams in class StatisticalModel

declareParameter

public void declareParameter(java.lang.String p)

Specified by:

declareParameter in interface DeclaredParameters

Overrides:

declareParameter in class StatisticalModel

declareParameter

public void declareParameter(java.lang.String p,
                    DblMatrix v)

Description copied from class: StatisticalModel

Convenience method that declares a parameter and sets a initial (default) value for that parameter.

Overrides:

declareParameter in class StatisticalModel

declareParameterSuper

public void declareParameterSuper(java.lang.String p)

declareParameterSuper

public void declareParameterSuper(java.lang.String p,
                         DblMatrix v)

revokeParameter

public void revokeParameter(java.lang.String p)

Specified by:

revokeParameter in interface DeclaredParameters

Overrides:

revokeParameter in class StatisticalModel

revokeParameterSuper

public void revokeParameterSuper(java.lang.String p)

getParams

public DblParamSet getParams()

Overrides:

getParams in class StatisticalModel

getDeclaredParameters

public java.util.Set getDeclaredParameters()

Specified by:

getDeclaredParameters in interface DeclaredParameters

Overrides:

getDeclaredParameters in class StatisticalModel

getParam

public DblMatrix getParam(java.lang.String name)

Specified by:

getParam in interface NamedParameters

Overrides:

getParam in class StatisticalModel

setParam

public void setParam(java.lang.String name,
            DblMatrix value)

Specified by:

setParam in interface NamedParameters

Overrides:

setParam in class StatisticalModel

replaceParams

public void replaceParams(DblParamSet X)

Description copied from class: StatisticalModel

Replace parameters with those in X. Note only the parameters in X will be modified. Any other parameters will remain unmodified.

Specified by:

replaceParams in interface NamedParameters

Overrides:

replaceParams in class StatisticalModel

hasParameter

public boolean hasParameter(java.lang.String s)

Specified by:

hasParameter in interface NamedParameters

Overrides:

hasParameter in class StatisticalModel

parameterSet

public java.lang.String[] parameterSet()

Specified by:

parameterSet in interface NamedParameters

Overrides:

parameterSet in class StatisticalModel

getParamSuper

public DblMatrix getParamSuper(java.lang.String name)

Provides access to the StatisticalModel super class method getParam. This allows some parameters to be stored as usual in a DblParamSet associated with the StatisticalModel.

setParamSuper

public void setParamSuper(java.lang.String name,
                 DblMatrix value)

Provides access to the StatisticalModel super class method setParam. This allows some parameters to be stored as usual in a DblParamSet associated with the StatisticalModel.

distributedNumParams

public int distributedNumParams()

declareDistributedParameter

protected abstract void declareDistributedParameter(java.lang.String p)

convertStringToDbl

public static DblMatrix convertStringToDbl(java.lang.String in)

convertDblToString

public static java.lang.String convertDblToString(DblMatrix in)

declareDistributedParameter

protected void declareDistributedParameter(java.lang.String p,
                               DblMatrix v)

declareIfNotDistributedParameter

protected boolean declareIfNotDistributedParameter(java.lang.String p,
                                       DblMatrix v)

revokeDistributedParameter

protected void revokeDistributedParameter(java.lang.String p)

getDistributedParams

public DblParamSet getDistributedParams()

getDeclaredDistributedParameters

public java.util.Set getDeclaredDistributedParameters()

getDistributedParam

public DblMatrix getDistributedParam(java.lang.String paramname)

setDistributedParam

public void setDistributedParam(java.lang.String paramname,
                       DblMatrix value)

replaceDistributedParams

public void replaceDistributedParams(DblParamSet X)

hasDistributedParameter

public boolean hasDistributedParameter(java.lang.String s)

distributedParameterSet

public java.lang.String[] distributedParameterSet()