sgpp::datadriven::DBMatOffline Class Reference

Class that is used to decompose and store the left-hand-side matrix for the density based classification approach (The classification is divided into two parts: the offline step that does not depend on the actual data and the online step that depends on the data). More...

#include <DBMatOffline.hpp>

Inheritance diagram for sgpp::datadriven::DBMatOffline:

Public Member Functions
virtual void	buildMatrix (Grid *grid, const RegularizationConfiguration &regularizationConfig)
	Builds the right hand side matrix with or without the regularization term depending on the type of decomposition.
virtual DBMatOffline *	clone () const =0
	Interface for the clone idiom.
virtual void	compute_inverse ()
virtual void	compute_inverse_parallel (std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)
	parallel/distributed version of compute_inverse()
void	compute_L2_refine_vectors (DataMatrix *mat_refine, Grid *grid, size_t newPoints)
	computes vectors of L2 products of grid points for refinement
	DBMatOffline (const DBMatOffline &rhs)
	Copy Constructor.
	DBMatOffline (const std::string &fileName)
	Constructor Create offline object from serialized offline object.
	DBMatOffline (DBMatOffline &&rhs)=default
	Default move constructor.
virtual void	decomposeMatrix (const RegularizationConfiguration &regularizationConfig, const DensityEstimationConfiguration &densityEstimationConfig)=0
	Decomposes the matrix according to the chosen decomposition type.
virtual void	decomposeMatrixParallel (RegularizationConfiguration &regularizationConfig, DensityEstimationConfiguration &densityEstimationConfig, std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)
	The parallel/distributed version of decomposeMatrix(...)
DataMatrixDistributed &	getDecomposedInverseDistributed ()
	Get a reference to the distributed decomposed matrix, analogously to getDecomposedMatrixDistributed()
DataMatrix &	getDecomposedMatrix ()
	Get a reference to the decomposed matrix.
DataMatrixDistributed &	getDecomposedMatrixDistributed ()
	Get a reference to the distributed decomposed matrix.
virtual sgpp::datadriven::MatrixDecompositionType	getDecompositionType ()=0
	Returns the decomposition type of the DBMatOffline object.
virtual size_t	getGridSize ()
	Returns the dimensionality of the quadratic lhs matrix (i.e.
DataMatrix &	getInverseMatrix ()
	Get a reference to the inverse matrix.
DataMatrix &	getLhsMatrix_ONLY_FOR_TESTING ()
	Allows access to lhs matrix, which is meant ONLY FOR TESTING.
virtual const DataMatrix &	getUnmodifiedR ()=0
	Get the unmodified (without added lambda) system matrix R.
virtual const DataMatrixDistributed &	getUnmodifiedRDistributed (std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)=0
	Get the distributed version of the unmodified (without added lambda) system matrix R.
virtual bool	isRefineable ()=0
	Only Offline objects based on Cholesky decomposition, or orthogonal adaptivity can be refined.
DBMatOffline &	operator= (const DBMatOffline &rhs)
	Default copy assign operator.
DBMatOffline &	operator= (DBMatOffline &&rhs)=default
	Default move assign operator.
void	printMatrix ()
	Prints the matrix onto standard output.
virtual void	store (const std::string &fileName)
	Serialize the DBMatOffline Object.
virtual void	syncDistributedDecomposition (std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)
	Synchronizes the decomposed matrix.
virtual void	syncDistributedInverse (std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)
	Synchronizes the inverse matrix.
virtual void	updateRegularization (double lambda)=0
	Modifies the decomposition to update the regularization parameter lambda.
virtual void	updateRegularizationParallel (double lambda, std::shared_ptr< BlacsProcessGrid > processGrid, const ParallelConfiguration &parallelConfig)=0
	Modifies the parallel decomposition to update the regularization parameter lambda.
virtual	~DBMatOffline ()=default
	Default virtual destructor.

Public Attributes
std::set< std::set< size_t > >	interactions

Protected Member Functions
	DBMatOffline ()
void	parseInter (const std::string &fileName, std::set< std::set< size_t > > &interactions) const
	Read the Interactionsterms from a serialized DBMatOfflibe object.

Protected Attributes
bool	isConstructed
bool	isDecomposed
DataMatrixDistributed	lhsDistributed
DataMatrixDistributed	lhsDistributedInverse
DataMatrix	lhsInverse
DataMatrix	lhsMatrix

Detailed Description

Class that is used to decompose and store the left-hand-side matrix for the density based classification approach (The classification is divided into two parts: the offline step that does not depend on the actual data and the online step that depends on the data).

Uses Gnu Scientific Library (GSL).

Constructor & Destructor Documentation

DBMatOffline() [1/4]

sgpp::datadriven::DBMatOffline::DBMatOffline ( const std::string &  fileName )

explicit

Constructor Create offline object from serialized offline object.

Parameters

fileName

path to the file that stores serialized offline object

References interactions, and parseInter().

DBMatOffline() [2/4]

sgpp::datadriven::DBMatOffline::DBMatOffline

(

const DBMatOffline &

rhs

)

Copy Constructor.

Parameters

rhs	Object to copy

DBMatOffline() [3/4]

sgpp::datadriven::DBMatOffline::DBMatOffline ( DBMatOffline &&  rhs )

default

Default move constructor.

~DBMatOffline()

virtual sgpp::datadriven::DBMatOffline::~DBMatOffline ( )

virtualdefault

Default virtual destructor.

DBMatOffline() [4/4]

sgpp::datadriven::DBMatOffline::DBMatOffline ( )

protected

Member Function Documentation

buildMatrix()

void sgpp::datadriven::DBMatOffline::buildMatrix ( Grid *  grid, const RegularizationConfiguration &  regularizationConfig  )

virtual

Builds the right hand side matrix with or without the regularization term depending on the type of decomposition.

Parameters

grid	The grid object the matrix is based on
regularizationConfig	Configures the regularization which is incorporated into the lhs

Reimplemented in sgpp::datadriven::DBMatOfflineGE, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

References sgpp::op_factory::createOperationLTwoDotExplicit(), grid(), isConstructed, and lhsMatrix.

Referenced by sgpp::datadriven::DBMatOfflineGE::buildMatrix(), sgpp::datadriven::DBMatOfflineOrthoAdapt::buildMatrix(), sgpp::datadriven::ModelFittingDensityEstimationOnOffParallel::fit(), sgpp::datadriven::DBMatPermutationFactory::getPermutedObject(), and main().

clone()

virtual DBMatOffline * sgpp::datadriven::DBMatOffline::clone ( ) const

pure virtual

Interface for the clone idiom.

Returns

a copy of this very object as a pointer to a new DBMatOffline object which is owned by the caller.

Implemented in sgpp::datadriven::DBMatOfflineChol, sgpp::datadriven::DBMatOfflineDenseIChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Referenced by sgpp::datadriven::ModelFittingDensityEstimationOnOff::fit(), and sgpp::datadriven::DBMatPermutationFactory::getPermutedObject().

compute_inverse()

void sgpp::datadriven::DBMatOffline::compute_inverse ( )

virtual

Reimplemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

compute_inverse_parallel()

void sgpp::datadriven::DBMatOffline::compute_inverse_parallel ( std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

virtual

parallel/distributed version of compute_inverse()

Parameters

processGrid	process grid to distribute the matrix on
parallelConfig

Reimplemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Referenced by sgpp::datadriven::ModelFittingDensityEstimationOnOffParallel::fit().

compute_L2_refine_vectors()

void sgpp::datadriven::DBMatOffline::compute_L2_refine_vectors	(	DataMatrix *	mat_refine,
		Grid *	grid,
		size_t	newPoints
	)

computes vectors of L2 products of grid points for refinement

Parameters

mat_refine	matrix to store L2 vectors
grid	underlying grid
newPoints	amount of points to refine

References grid(), sgpp::base::Linear, and sgpp::base::ModLinear.

Referenced by sgpp::datadriven::DBMatOfflineChol::choleskyModification().

decomposeMatrix()

virtual void sgpp::datadriven::DBMatOffline::decomposeMatrix ( const RegularizationConfiguration &  regularizationConfig, const DensityEstimationConfiguration &  densityEstimationConfig  )

pure virtual

Decomposes the matrix according to the chosen decomposition type.

The number of rows of the stored result depends on the decomposition type.

Parameters

regularizationConfig	the regularization configuration
densityEstimationConfig	the density estimation configuration

Implemented in sgpp::datadriven::DBMatOfflineChol, sgpp::datadriven::DBMatOfflineDenseIChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Referenced by sgpp::datadriven::ModelFittingDensityEstimationOnOffParallel::fit(), sgpp::datadriven::DBMatPermutationFactory::getPermutedObject(), and main().

decomposeMatrixParallel()

void sgpp::datadriven::DBMatOffline::decomposeMatrixParallel ( RegularizationConfiguration &  regularizationConfig, DensityEstimationConfiguration &  densityEstimationConfig, std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

virtual

The parallel/distributed version of decomposeMatrix(...)

Parameters

regularizationConfig	the regularization configuration
densityEstimationConfig	the density estimation configuration
processGrid	process grid to distribute the matrix on
parallelConfig

Reimplemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Referenced by sgpp::datadriven::ModelFittingDensityEstimationOnOffParallel::fit().

getDecomposedInverseDistributed()

DataMatrixDistributed & sgpp::datadriven::DBMatOffline::getDecomposedInverseDistributed

(

)

Get a reference to the distributed decomposed matrix, analogously to getDecomposedMatrixDistributed()

References isDecomposed, and lhsDistributedInverse.

Referenced by sgpp::datadriven::DBMatOnlineDE_SMW::smw_adapt_parallel(), and sgpp::datadriven::DBMatOnlineDE_SMW::solveSLEParallel().

getDecomposedMatrix()

DataMatrix & sgpp::datadriven::DBMatOffline::getDecomposedMatrix

(

)

Get a reference to the decomposed matrix.

Throws if matrix has not yet been decomposed.

Returns

decomposed matrix

References isDecomposed, and lhsMatrix.

Referenced by sgpp::datadriven::DBMatOnlineDE::computeDensityDifferenceFunction(), sgpp::datadriven::DBMatOnlineDE::computeDensityFunction(), sgpp::datadriven::DBMatOnlineDE::computeDensityFunctionParallel(), sgpp::datadriven::LearnerSGDEOnOffParallel::computeNewSystemMatrixDecomposition(), sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatch(), sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatchParallel(), sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatchTwoDatasets(), and sgpp::datadriven::DBMatOnlineDEChol::solveSLE().

getDecomposedMatrixDistributed()

DataMatrixDistributed & sgpp::datadriven::DBMatOffline::getDecomposedMatrixDistributed

(

)

Get a reference to the distributed decomposed matrix.

Throws if matrix has not yet been decomposed. In order to return valid data, syncDistributedDecomposition() has to be called if the decomposition was changed (after refinements).

References isDecomposed, and lhsDistributed.

Referenced by sgpp::datadriven::DBMatOnlineDEChol::solveSLEParallel().

getDecompositionType()

virtual sgpp::datadriven::MatrixDecompositionType sgpp::datadriven::DBMatOffline::getDecompositionType ( )

pure virtual

Returns the decomposition type of the DBMatOffline object.

Returns

the type of matrix decomposition

Implemented in sgpp::datadriven::DBMatOfflineChol, sgpp::datadriven::DBMatOfflineDenseIChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Referenced by sgpp::datadriven::DBMatOnlineDEChol::buildCholSolver(), sgpp::datadriven::DBMatOnlineDE_SMW::DBMatOnlineDE_SMW(), sgpp::datadriven::DBMatOnlineDEOrthoAdapt::DBMatOnlineDEOrthoAdapt(), and store().

getGridSize()

size_t sgpp::datadriven::DBMatOffline::getGridSize ( )

virtual

Returns the dimensionality of the quadratic lhs matrix (i.e.

the number of rows)

Returns

the grid size

References sgpp::base::DataMatrix::getNrows(), and lhsMatrix.

Referenced by python.learner.TrainingStopPolicy.TrainingStopPolicy::hasGridSizeChanged(), python.learner.TrainingStopPolicy.TrainingStopPolicy::isTrainingComplete(), sgpp::datadriven::DBMatOnlineDEOrthoAdapt::sherman_morrison_adapt(), sgpp::datadriven::DBMatOnlineDE_SMW::smw_adapt(), sgpp::datadriven::DBMatOnlineDE_SMW::smw_adapt_parallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::store(), sgpp::datadriven::DBMatOnlineDE_SMW::updateSystemMatrixDecomposition(), sgpp::datadriven::DBMatOnlineDEOrthoAdapt::updateSystemMatrixDecomposition(), and sgpp::datadriven::DBMatOnlineDE_SMW::updateSystemMatrixDecompositionParallel().

getInverseMatrix()

DataMatrix & sgpp::datadriven::DBMatOffline::getInverseMatrix

(

)

Get a reference to the inverse matrix.

Returns

inverse matrix

References lhsInverse.

Referenced by sgpp::datadriven::DBMatOnlineDE_SMW::smw_adapt(), and sgpp::datadriven::DBMatOnlineDE_SMW::solveSLE().

getLhsMatrix_ONLY_FOR_TESTING()

sgpp::base::DataMatrix & sgpp::datadriven::DBMatOffline::getLhsMatrix_ONLY_FOR_TESTING

(

)

Allows access to lhs matrix, which is meant ONLY FOR TESTING.

References lhsMatrix.

getUnmodifiedR()

virtual const DataMatrix & sgpp::datadriven::DBMatOffline::getUnmodifiedR ( )

pure virtual

Get the unmodified (without added lambda) system matrix R.

Returns

Matrix R

Implemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

getUnmodifiedRDistributed()

virtual const DataMatrixDistributed & sgpp::datadriven::DBMatOffline::getUnmodifiedRDistributed ( std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

pure virtual

Get the distributed version of the unmodified (without added lambda) system matrix R.

Only possible in the ScaLAPACK version.

Parameters

processGrid	BLACS process grid
parallelConfig	config options for ScaLAPACK

Returns

Matrix R

Implemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

isRefineable()

virtual bool sgpp::datadriven::DBMatOffline::isRefineable ( )

pure virtual

Only Offline objects based on Cholesky decomposition, or orthogonal adaptivity can be refined.

Returns

true if object can be refined, else false;

Implemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

operator=() [1/2]

DBMatOffline & sgpp::datadriven::DBMatOffline::operator=

(

const DBMatOffline &

rhs

)

Default copy assign operator.

References interactions, isConstructed, isDecomposed, lhsInverse, and lhsMatrix.

operator=() [2/2]

DBMatOffline & sgpp::datadriven::DBMatOffline::operator= ( DBMatOffline &&  rhs )

default

Default move assign operator.

parseInter()

void sgpp::datadriven::DBMatOffline::parseInter ( const std::string &  fileName, std::set< std::set< size_t > > &  interactions  ) const

protected

Read the Interactionsterms from a serialized DBMatOfflibe object.

Parameters

fileName	path of the serialized DBMatOffline object
interactions	the interactions to populate

References fileName, and sgpp::base::StringTokenizer::tokenize().

Referenced by DBMatOffline().

printMatrix()

void sgpp::datadriven::DBMatOffline::printMatrix

(

)

Prints the matrix onto standard output.

References sgpp::base::DataMatrix::getNcols(), sgpp::base::DataMatrix::getNrows(), isDecomposed, lhsMatrix, and sgpp::base::DataMatrix::toString().

store()

void sgpp::datadriven::DBMatOffline::store ( const std::string &  fileName )

virtual

Serialize the DBMatOffline Object.

Parameters

fileName

path where to store the file.

Reimplemented in sgpp::datadriven::DBMatOfflineOrthoAdapt.

References fileName, getDecompositionType(), sgpp::base::DataMatrix::getNcols(), sgpp::base::DataMatrix::getNrows(), sgpp::base::DataMatrix::getPointer(), interactions, isDecomposed, and lhsMatrix.

Referenced by main(), and sgpp::datadriven::DBMatOfflineOrthoAdapt::store().

syncDistributedDecomposition()

void sgpp::datadriven::DBMatOffline::syncDistributedDecomposition ( std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

virtual

Synchronizes the decomposed matrix.

Override if more matrices have to be synched.

Parameters

processGrid	process grid to distribute the matrix on
parallelConfig

Reimplemented in sgpp::datadriven::DBMatOfflineOrthoAdapt.

References sgpp::datadriven::ParallelConfiguration::columnBlockSize_, sgpp::datadriven::DataMatrixDistributed::fromSharedData(), sgpp::base::DataMatrix::getNcols(), sgpp::base::DataMatrix::getNrows(), isDecomposed, lhsDistributed, lhsMatrix, and sgpp::datadriven::ParallelConfiguration::rowBlockSize_.

Referenced by sgpp::datadriven::DBMatOnlineDE::syncDistributedDecomposition(), and sgpp::datadriven::DBMatOnlineDEOrthoAdapt::syncDistributedDecomposition().

syncDistributedInverse()

void sgpp::datadriven::DBMatOffline::syncDistributedInverse ( std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

virtual

Synchronizes the inverse matrix.

Parameters

processGrid	process grid to distribute the matrix on
parallelConfig

References sgpp::datadriven::ParallelConfiguration::columnBlockSize_, sgpp::datadriven::DataMatrixDistributed::fromSharedData(), sgpp::base::DataMatrix::getNcols(), sgpp::base::DataMatrix::getNrows(), isDecomposed, lhsDistributedInverse, lhsInverse, and sgpp::datadriven::ParallelConfiguration::rowBlockSize_.

updateRegularization()

virtual void sgpp::datadriven::DBMatOffline::updateRegularization ( double  lambda )

pure virtual

Modifies the decomposition to update the regularization parameter lambda.

Parameters

lambda

New lambda value

Implemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

updateRegularizationParallel()

virtual void sgpp::datadriven::DBMatOffline::updateRegularizationParallel ( double  lambda, std::shared_ptr< BlacsProcessGrid >  processGrid, const ParallelConfiguration &  parallelConfig  )

pure virtual

Modifies the parallel decomposition to update the regularization parameter lambda.

Parameters

lambda	New lambda value
processGrid	ScaLAPACK process grid
parallelConfig	Configuration for ScaLAPACK

Implemented in sgpp::datadriven::DBMatOfflineChol, and sgpp::datadriven::DBMatOfflineOrthoAdapt.

Member Data Documentation

interactions

std::set<std::set<size_t> > sgpp::datadriven::DBMatOffline::interactions

Referenced by sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatch(), sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatchParallel(), sgpp::datadriven::DBMatOnlineDE::computeWeightedBFromBatchTwoDatasets(), DBMatOffline(), sgpp::datadriven::DBMatOnlineDE::eval(), sgpp::datadriven::DBMatOnlineDE::evalParallel(), operator=(), and store().

isConstructed

bool sgpp::datadriven::DBMatOffline::isConstructed

protected

Referenced by buildMatrix(), sgpp::datadriven::DBMatOfflineGE::buildMatrix(), sgpp::datadriven::DBMatOfflineOrthoAdapt::DBMatOfflineOrthoAdapt(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineDenseIChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrix(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrixParallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrixParallel(), and operator=().

isDecomposed

bool sgpp::datadriven::DBMatOffline::isDecomposed

protected

Referenced by sgpp::datadriven::DBMatOfflineChol::choleskyAddPoint(), sgpp::datadriven::DBMatOfflineChol::choleskyPermutation(), sgpp::datadriven::DBMatOfflineChol::compute_inverse(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse(), sgpp::datadriven::DBMatOfflineChol::compute_inverse_parallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse_parallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::DBMatOfflineOrthoAdapt(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineDenseIChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrix(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrixParallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrixParallel(), getDecomposedInverseDistributed(), getDecomposedMatrix(), getDecomposedMatrixDistributed(), operator=(), printMatrix(), store(), syncDistributedDecomposition(), sgpp::datadriven::DBMatOfflineOrthoAdapt::syncDistributedDecomposition(), and syncDistributedInverse().

lhsDistributed

DataMatrixDistributed sgpp::datadriven::DBMatOffline::lhsDistributed

protected

Referenced by sgpp::datadriven::DBMatOfflineChol::decomposeMatrixParallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrixParallel(), getDecomposedMatrixDistributed(), sgpp::datadriven::DBMatOfflineOrthoAdapt::getUnmodifiedRDistributed(), and syncDistributedDecomposition().

lhsDistributedInverse

DataMatrixDistributed sgpp::datadriven::DBMatOffline::lhsDistributedInverse

protected

Referenced by sgpp::datadriven::DBMatOfflineChol::compute_inverse_parallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse_parallel(), getDecomposedInverseDistributed(), and syncDistributedInverse().

lhsInverse

DataMatrix sgpp::datadriven::DBMatOffline::lhsInverse

protected

Referenced by sgpp::datadriven::DBMatOfflineChol::compute_inverse(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse(), sgpp::datadriven::DBMatOfflineChol::compute_inverse_parallel(), getInverseMatrix(), operator=(), and syncDistributedInverse().

lhsMatrix

DataMatrix sgpp::datadriven::DBMatOffline::lhsMatrix

protected

Referenced by buildMatrix(), sgpp::datadriven::DBMatOfflineGE::buildMatrix(), sgpp::datadriven::DBMatOfflineChol::choleskyAddPoint(), sgpp::datadriven::DBMatOfflineChol::choleskyModification(), sgpp::datadriven::DBMatOfflineDenseIChol::choleskyModification(), sgpp::datadriven::DBMatOfflineChol::choleskyPermutation(), sgpp::datadriven::DBMatOfflineChol::compute_inverse(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse(), sgpp::datadriven::DBMatOfflineChol::compute_inverse_parallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::compute_inverse_parallel(), sgpp::datadriven::DBMatOfflineGE::DBMatOfflineGE(), sgpp::datadriven::DBMatOfflineOrthoAdapt::DBMatOfflineOrthoAdapt(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineDenseIChol::decomposeMatrix(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrix(), sgpp::datadriven::DBMatOfflineChol::decomposeMatrixParallel(), sgpp::datadriven::DBMatOfflineOrthoAdapt::decomposeMatrixParallel(), getDecomposedMatrix(), getGridSize(), getLhsMatrix_ONLY_FOR_TESTING(), sgpp::datadriven::DBMatOfflineOrthoAdapt::getUnmodifiedR(), sgpp::datadriven::DBMatOfflineOrthoAdapt::getUnmodifiedRDistributed(), sgpp::datadriven::DBMatOfflineOrthoAdapt::hessenberg_decomposition(), sgpp::datadriven::DBMatOfflineOrthoAdapt::invert_symmetric_tridiag(), operator=(), sgpp::datadriven::DBMatOfflinePermutable::permuteLhsMatrix(), printMatrix(), store(), syncDistributedDecomposition(), sgpp::datadriven::DBMatOfflineOrthoAdapt::updateRegularization(), and sgpp::datadriven::DBMatOfflineOrthoAdapt::updateRegularizationParallel().

---

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

• datadriven/src/sgpp/datadriven/algorithm/DBMatOffline.hpp
• datadriven/src/sgpp/datadriven/algorithm/DBMatOffline.cpp