Abstract super class to encapsulate density estimation models working with multiple Grids. More...

#include <ModelFittingDensityEstimationCombi.hpp>

Inheritance diagram for sgpp::datadriven::ModelFittingDensityEstimationCombi:

Public Member Functions
bool	adapt () override
	Refines the component with the biggest error.

bool	adapt (size_t newNoPoints, std::vector< size_t > &deletedGridPoints) override
	Currently not implemented for this class due to missing strategy for dimensional adaptive refinement.

double	computeResidual (DataMatrix &validationData) const override
	Should compute some kind of Residual to evaluate the fit of the model.

double	evaluate (const DataVector &sample) override
	Evaluate the fitted density at a single data point - requires a trained grid.

void	evaluate (DataMatrix &samples, DataVector &results) override
	Evaluate the fitted density on a set of data points - requires a trained grid.

void	fit (DataMatrix &newDataset) override
	Fit the grids to the given dataset by determining the surpluses of the initial grid by the SGDE approach.

void	fit (DataMatrix &newDatasetP, DataMatrix &newDatasetQ) override

void	fit (Dataset &newDataset) override
	Fit the grids to the given dataset by determining the weights of the initial grid by the SGDE approach.

void	fit (Dataset &newDatasetP, Dataset &newDatasetQ) override

	ModelFittingDensityEstimationCombi ()
	Default constructor.

	ModelFittingDensityEstimationCombi (const FitterConfigurationDensityEstimation &config)
	Constructor from a configuration object generated by the datamining pipeline.

	ModelFittingDensityEstimationCombi (const FitterConfigurationDensityEstimation &config, std::shared_ptr< DBMatObjectStore > objectStore)
	Construct from a configuration object generated from the datamining pipeline and a object store to obtain and store already decomposed offline objects.

void	reset () override
	Resets the state of the entire model.

void	resetTraining () override
	Resets any trained representations of the model, but does not reset the entire state.

void	update (DataMatrix &samples) override
	Updates the model based on new data samples (streaming, batch learning).

void	update (DataMatrix &samplesP, DataMatrix &samplesQ) override

void	update (Dataset &dataset) override
	Train the grid of an existing model with new samples.

void	update (Dataset &datasetP, Dataset &datasetQ) override

void	updateRegularization (double lambda) override
	Updates the regularization parameter lambda of the underlying model.

Public Member Functions inherited from sgpp::datadriven::ModelFittingDensityEstimation
	ModelFittingDensityEstimation ()
	Default constructor.

Public Member Functions inherited from sgpp::datadriven::ModelFittingBaseSingleGrid
Grid &	getGrid ()
	Get the underlying grid object for the current model.

DataVector &	getSurpluses ()
	Get the surpluses of the current grid.

	ModelFittingBaseSingleGrid ()
	Default constructor.

	ModelFittingBaseSingleGrid (const ModelFittingBaseSingleGrid &rhs)=delete
	Copy constructor - we cannot deep copy all member variables yet.

	ModelFittingBaseSingleGrid (ModelFittingBaseSingleGrid &&rhs)=default
	Move constructor.

ModelFittingBaseSingleGrid &	operator= (const ModelFittingBaseSingleGrid &rhs)=delete
	Copy assign operator - we cannot deep copy all member variables yet.

ModelFittingBaseSingleGrid &	operator= (ModelFittingBaseSingleGrid &&rhs)=default
	Move assign operator.

std::string	storeFitter ()

virtual	~ModelFittingBaseSingleGrid ()=default
	virtual destructor.

Public Member Functions inherited from sgpp::datadriven::ModelFittingBase
Dataset *	getDataset ()

FitterConfiguration &	getFitterConfiguration ()
	Get or set the configuration of the fitter object.

const FitterConfiguration &	getFitterConfiguration () const
	Get the configuration of the fitter object.

virtual std::shared_ptr< BlacsProcessGrid >	getProcessGrid () const

	ModelFittingBase ()
	Default constructor.

	ModelFittingBase (const ModelFittingBase &rhs)=delete
	Copy constructor - we cannot deep copy all member variables yet.

	ModelFittingBase (ModelFittingBase &&rhs)=default
	Move constructor.

ModelFittingBase &	operator= (const ModelFittingBase &rhs)=delete
	Copy assign operator - we cannot deep copy all member variables yet.

ModelFittingBase &	operator= (ModelFittingBase &&rhs)=default
	Move assign operator.

virtual	~ModelFittingBase ()=default
	virtual destructor.

Protected Member Functions
void	addNewModel (const std::pair< std::vector< size_t >, int > combiconfig)

std::unique_ptr< ModelFittingDensityEstimation >	createNewModel (sgpp::datadriven::FitterConfigurationDensityEstimation &densityEstimationConfig)
	Creates a density estimation model that fits the model settings.

bool	isRefinable () override
	Function that indicates whether a model is refinable at all (certain on/off settings do not allow for refinement)

void	removeModel (size_t indexRev)

Protected Member Functions inherited from sgpp::datadriven::ModelFittingDensityEstimation
std::unique_ptr< sgpp::base::CoarseningFunctor >	getCoarseningFunctor ()
	Returns the refinement functor suitable for the model settings.

std::unique_ptr< sgpp::base::RefinementFunctor >	getRefinementFunctor ()
	Returns the refinement functor suitable for the model settings.

Protected Member Functions inherited from sgpp::datadriven::ModelFittingBase
Grid *	buildGrid (const sgpp::base::GeneralGridConfiguration &gridConfig) const
	Factory member function that generates a grid from configuration.

Grid *	buildGrid (const sgpp::base::GeneralGridConfiguration &gridConfig, const GeometryConfiguration &geometryConfig) const
	Factory member function that generates a grid from configuration.

SLESolver *	buildSolver (const SLESolverConfiguration &config) const
	Factory member function to build the solver for the least squares regression problem according to the config.

std::set< std::set< size_t > >	getInteractions (const GeometryConfiguration &geometryConfig)

void	reconfigureSolver (SLESolver &solver, const SLESolverConfiguration &config) const
	Configure solver based on the desired configuration.

Protected Attributes
std::vector< std::pair< std::vector< size_t >, int > >	componentConfigs
	Contains all level vector and weights of the current component grid set.

std::vector< std::unique_ptr< ModelFittingDensityEstimation > >	components
	Contains the component grids witch form the sparse grids.

std::vector< bool >	fitted
	Contains the status of the component grids.

bool	hasObjectStore
	True if the instance has an object store.

std::shared_ptr< DBMatObjectStore >	objectStore
	The object store passed to component objects.

CombiScheme	scheme
	Delivers the initial level vectors and weighs and manages refinements.

Protected Attributes inherited from sgpp::datadriven::ModelFittingDensityEstimation
size_t	initialGridSize
	Initial number of grid points.

size_t	refinementsPerformed
	Count the amount of refinement operations performed on the current dataset.

Protected Attributes inherited from sgpp::datadriven::ModelFittingBaseSingleGrid
DataVector	alpha
	hierarchical surpluses of the grid.

std::unique_ptr< Grid >	grid
	the sparse grid that approximates the data.

Protected Attributes inherited from sgpp::datadriven::ModelFittingBase
std::unique_ptr< FitterConfiguration >	config
	Configuration object for the fitter.

Dataset *	dataset
	Pointer to sgpp::datadriven::Dataset.

Dataset *	extraDataset

std::unique_ptr< std::set< std::set< size_t > > >	interactions

std::unique_ptr< SLESolver >	solver
	Solver for the learning problem.

Additional Inherited Members
Public Attributes inherited from sgpp::datadriven::ModelFittingBase
bool	verboseSolver
	Whether the Solver produces output or not.

Detailed Description

Abstract super class to encapsulate density estimation models working with multiple Grids.

Constructor & Destructor Documentation

◆ ModelFittingDensityEstimationCombi() [1/3]

sgpp::datadriven::ModelFittingDensityEstimationCombi::ModelFittingDensityEstimationCombi ( )

Default constructor.

◆ ModelFittingDensityEstimationCombi() [2/3]

sgpp::datadriven::ModelFittingDensityEstimationCombi::ModelFittingDensityEstimationCombi ( const FitterConfigurationDensityEstimation & config )

explicit

Constructor from a configuration object generated by the datamining pipeline.

Parameters

config configuration object generated by the datamining pipeline

References components, sgpp::datadriven::ModelFittingBase::config, fitted, hasObjectStore, objectStore, and sgpp::datadriven::DBMatOfflinePermutable::PermutableDecompositions.

◆ ModelFittingDensityEstimationCombi() [3/3]

sgpp::datadriven::ModelFittingDensityEstimationCombi::ModelFittingDensityEstimationCombi	(	const FitterConfigurationDensityEstimation &	config,
		std::shared_ptr< DBMatObjectStore >	objectStore
	)

explicit

Construct from a configuration object generated from the datamining pipeline and a object store to obtain and store already decomposed offline objects.

Parameters

config	Configuration object generated by the datamining pipeline.
objectStore	Offline object store.

References hasObjectStore, and objectStore.

Member Function Documentation

◆ adapt() [1/2]

bool sgpp::datadriven::ModelFittingDensityEstimationCombi::adapt ( )

overridevirtual

Refines the component with the biggest error.

Returns: if a component was refined

Reimplemented from sgpp::datadriven::ModelFittingDensityEstimation.

References addNewModel(), componentConfigs, components, sgpp::datadriven::CombiScheme::getCombiScheme(), sgpp::datadriven::CombiScheme::isRefinable(), sgpp::datadriven::CombiScheme::refineComponent(), sgpp::datadriven::ModelFittingDensityEstimation::refinementsPerformed, removeModel(), and scheme.

◆ adapt() [2/2]

bool sgpp::datadriven::ModelFittingDensityEstimationCombi::adapt	(	size_t	newNoPoints,
		std::vector< size_t > &	deletedGridPoints
	)

overridevirtual

Currently not implemented for this class due to missing strategy for dimensional adaptive refinement.

Throws an application_exception.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ addNewModel()

void sgpp::datadriven::ModelFittingDensityEstimationCombi::addNewModel ( const std::pair< std::vector< size_t >, int > combiconfig )

protected

References componentConfigs, components, sgpp::datadriven::ModelFittingBase::config, createNewModel(), fitted, and sgpp::datadriven::FitterConfigurationDensityEstimation::setupDefaults().

Referenced by adapt().

◆ computeResidual()

double sgpp::datadriven::ModelFittingDensityEstimationCombi::computeResidual ( DataMatrix & validationData ) const

inlineoverridevirtual

Should compute some kind of Residual to evaluate the fit of the model.

In the case of density estimation, this is || R * alpha_lambda - b_val ||_2

This is useful for unsupervised learning models, where normal evaluation cannot be used as there are no targets.

Parameters

validationData Matrix for validation data

Returns: the residual score

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ createNewModel()

std::unique_ptr< ModelFittingDensityEstimation > sgpp::datadriven::ModelFittingDensityEstimationCombi::createNewModel ( sgpp::datadriven::FitterConfigurationDensityEstimation & densityEstimationConfig )

protected

Creates a density estimation model that fits the model settings.

Parameters

densityEstimationConfig configuration for the density estimation

Returns: a new density estimation model

References sgpp::datadriven::CG, sgpp::datadriven::Decomposition, sgpp::datadriven::FitterConfiguration::getDensityEstimationConfig(), hasObjectStore, objectStore, and sgpp::datadriven::DensityEstimationConfiguration::type_.

Referenced by addNewModel(), and fit().

◆ evaluate() [1/2]

double sgpp::datadriven::ModelFittingDensityEstimationCombi::evaluate ( const DataVector & sample )

overridevirtual

Evaluate the fitted density at a single data point - requires a trained grid.

Parameters

sample vector with the coordinates in all dimensions of that sample.

Returns: evaluation of the trained grid.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References componentConfigs, components, and fitted.

◆ evaluate() [2/2]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::evaluate	(	DataMatrix &	samples,
		DataVector &	results
	)

overridevirtual

Evaluate the fitted density on a set of data points - requires a trained grid.

Parameters

samples	matrix where each row represents a sample and the columns contain the coordinates in all dimensions of that sample.
results	vector where each row will contain the evaluation of the respective sample on the current model.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References sgpp::base::DataVector::add(), componentConfigs, components, fitted, and sgpp::base::DataVector::setAll().

◆ fit() [1/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::fit ( DataMatrix & newDataset )

overridevirtual

Fit the grids to the given dataset by determining the surpluses of the initial grid by the SGDE approach.

Requires only data samples and no targets (since those are irrelevant for the density estimation whatsoever)

Parameters

newDataset the training dataset that is used to fit the model.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References componentConfigs, sgpp::base::ComponentGrid, components, sgpp::datadriven::ModelFittingBase::config, createNewModel(), sgpp::datadriven::DensityEstimationConfiguration::decomposition_, fitted, sgpp::datadriven::CombiScheme::getCombiScheme(), sgpp::datadriven::FitterConfiguration::getDensityEstimationConfig(), sgpp::datadriven::ModelFittingBase::getFitterConfiguration(), sgpp::base::DataMatrix::getNcols(), sgpp::datadriven::FitterConfiguration::getRegularizationConfig(), sgpp::datadriven::CombiScheme::initialize(), sgpp::datadriven::RegularizationConfiguration::lambda_, scheme, and sgpp::datadriven::FitterConfigurationDensityEstimation::setupDefaults().

◆ fit() [2/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::fit	(	DataMatrix &	newDatasetP,
		DataMatrix &	newDatasetQ
	)

inlineoverridevirtual

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ fit() [3/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::fit ( Dataset & newDataset )

overridevirtual

Fit the grids to the given dataset by determining the weights of the initial grid by the SGDE approach.

Parameters

newDataset the training dataset that is used to fit the model.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References sgpp::datadriven::ModelFittingBase::dataset, fit(), and sgpp::datadriven::Dataset::getData().

Referenced by fit(), update(), and update().

◆ fit() [4/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::fit	(	Dataset &	newDatasetP,
		Dataset &	newDatasetQ
	)

inlineoverridevirtual

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ isRefinable()

bool sgpp::datadriven::ModelFittingDensityEstimationCombi::isRefinable ( )

overrideprotectedvirtual

Function that indicates whether a model is refinable at all (certain on/off settings do not allow for refinement)

Returns: whether the model is refinable

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ removeModel()

void sgpp::datadriven::ModelFittingDensityEstimationCombi::removeModel ( size_t indexRev )

protected

Parameters

indexRev the reverse index (distance from the end of the vector) of the component that must be removed

References componentConfigs, components, and fitted.

Referenced by adapt().

◆ reset()

void sgpp::datadriven::ModelFittingDensityEstimationCombi::reset ( )

overridevirtual

Resets the state of the entire model.

Implements sgpp::datadriven::ModelFittingBase.

References components, and sgpp::datadriven::ModelFittingDensityEstimation::refinementsPerformed.

◆ resetTraining()

void sgpp::datadriven::ModelFittingDensityEstimationCombi::resetTraining ( )

inlineoverridevirtual

Resets any trained representations of the model, but does not reset the entire state.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

◆ update() [1/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::update ( DataMatrix & samples )

overridevirtual

Updates the model based on new data samples (streaming, batch learning).

Requires only the data samples and no targets (since those are irrelevant for the density estimation whatsoever)

Parameters

samples the new data samples

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References components, fit(), fitted, and sgpp::datadriven::ModelFittingDensityEstimation::refinementsPerformed.

Referenced by python.uq.refinement.RefinementStrategy.Ranking::rank(), and python.learner.LearnedKnowledge.LearnedKnowledge::setMemento().

◆ update() [2/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::update	(	DataMatrix &	samplesP,
		DataMatrix &	samplesQ
	)

inlineoverridevirtual

Implements sgpp::datadriven::ModelFittingDensityEstimation.

Referenced by python.uq.refinement.RefinementStrategy.Ranking::rank(), and python.learner.LearnedKnowledge.LearnedKnowledge::setMemento().

◆ update() [3/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::update ( Dataset & dataset )

overridevirtual

Train the grid of an existing model with new samples.

Parameters

dataset the training dataset that is used to fit the model.

Implements sgpp::datadriven::ModelFittingDensityEstimation.

References components, sgpp::datadriven::ModelFittingBase::dataset, fit(), fitted, sgpp::datadriven::Dataset::getData(), and sgpp::datadriven::ModelFittingDensityEstimation::refinementsPerformed.

Referenced by python.uq.refinement.RefinementStrategy.Ranking::rank(), and python.learner.LearnedKnowledge.LearnedKnowledge::setMemento().

◆ update() [4/4]

void sgpp::datadriven::ModelFittingDensityEstimationCombi::update	(	Dataset &	datasetP,
		Dataset &	datasetQ
	)

inlineoverridevirtual

Implements sgpp::datadriven::ModelFittingDensityEstimation.

Referenced by python.uq.refinement.RefinementStrategy.Ranking::rank(), and python.learner.LearnedKnowledge.LearnedKnowledge::setMemento().

◆ updateRegularization()

void sgpp::datadriven::ModelFittingDensityEstimationCombi::updateRegularization ( double lambda )

inlineoverridevirtual

Updates the regularization parameter lambda of the underlying model.

Parameters

lambda the new lambda parameter

Implements sgpp::datadriven::ModelFittingDensityEstimation.

Member Data Documentation

◆ componentConfigs

std::vector<std::pair<std::vector<size_t>, int> > sgpp::datadriven::ModelFittingDensityEstimationCombi::componentConfigs

protected

Contains all level vector and weights of the current component grid set.

Referenced by adapt(), addNewModel(), evaluate(), evaluate(), fit(), and removeModel().

◆ components

std::vector<std::unique_ptr<ModelFittingDensityEstimation> > sgpp::datadriven::ModelFittingDensityEstimationCombi::components

protected

Contains the component grids witch form the sparse grids.

Referenced by adapt(), addNewModel(), evaluate(), evaluate(), fit(), ModelFittingDensityEstimationCombi(), removeModel(), reset(), update(), and update().

◆ fitted

std::vector<bool> sgpp::datadriven::ModelFittingDensityEstimationCombi::fitted

protected

Contains the status of the component grids.

true: fitted false: unfitted

Referenced by addNewModel(), evaluate(), evaluate(), fit(), ModelFittingDensityEstimationCombi(), removeModel(), update(), and update().

◆ hasObjectStore

bool sgpp::datadriven::ModelFittingDensityEstimationCombi::hasObjectStore

protected

True if the instance has an object store.

Referenced by createNewModel(), ModelFittingDensityEstimationCombi(), and ModelFittingDensityEstimationCombi().

◆ objectStore

std::shared_ptr<DBMatObjectStore> sgpp::datadriven::ModelFittingDensityEstimationCombi::objectStore

protected

The object store passed to component objects.

Referenced by createNewModel(), ModelFittingDensityEstimationCombi(), and ModelFittingDensityEstimationCombi().

◆ scheme

CombiScheme sgpp::datadriven::ModelFittingDensityEstimationCombi::scheme

protected

Delivers the initial level vectors and weighs and manages refinements.

Referenced by adapt(), and fit().

The documentation for this class was generated from the following files:

datadriven/src/sgpp/datadriven/datamining/modules/fitting/ModelFittingDensityEstimationCombi.hpp
datadriven/src/sgpp/datadriven/datamining/modules/fitting/ModelFittingDensityEstimationCombi.cpp

Public Member Functions

Protected Member Functions

Protected Attributes

Additional Inherited Members

Detailed Description

Constructor & Destructor Documentation

◆ ModelFittingDensityEstimationCombi() [1/3]

◆ ModelFittingDensityEstimationCombi() [2/3]

◆ ModelFittingDensityEstimationCombi() [3/3]

Member Function Documentation

◆ adapt() [1/2]

◆ adapt() [2/2]

◆ addNewModel()

◆ computeResidual()

◆ createNewModel()

◆ evaluate() [1/2]

◆ evaluate() [2/2]

◆ fit() [1/4]

◆ fit() [2/4]

◆ fit() [3/4]

◆ fit() [4/4]

◆ isRefinable()

◆ removeModel()

◆ reset()

◆ resetTraining()

◆ update() [1/4]

◆ update() [2/4]

◆ update() [3/4]

◆ update() [4/4]

◆ updateRegularization()

Member Data Documentation

◆ componentConfigs

◆ components

◆ fitted

◆ hasObjectStore

◆ objectStore

◆ scheme