sgpp::datadriven::DMSystemMatrixBase Class Reference

Abstract class that defines the virtual class base::OperationMatrix for classification and regression problems. More...

#include <DMSystemMatrixBase.hpp>

Inheritance diagram for sgpp::datadriven::DMSystemMatrixBase:

Public Member Functions
	DMSystemMatrixBase (base::DataMatrix &trainData, double lambda)
	Std-Constructor. More...
virtual void	generateb (base::DataVector &classes, base::DataVector &b)=0
	Generates the right hand side of the classification equation. More...
virtual void	getTimers (double &timeMult, double &computeMult, double &timeMultTrans, double &computeMultTrans)
	gets the timer's values by saving them into call by reference values More...
virtual void	mult (base::DataVector &alpha, base::DataVector &result)=0
	starts the Multiplication with the matrix More...
virtual void	prepareGrid ()
	forward declaration More...
virtual void	resetTimers ()
	resets all timers to 0 More...
virtual	~DMSystemMatrixBase ()
	Std-Destructor. More...
Public Member Functions inherited from sgpp::base::OperationMatrix
	OperationMatrix ()
	Constructor. More...
virtual	~OperationMatrix ()
	Destructor. More...

Protected Attributes
double	completeTimeMult_
	time needed for Mult More...
double	completeTimeMultTrans_
	time needed for Mult transposed More...
double	computeTimeMult_
	time needed only for the computation of mult, interesting on accelerator boards More...
double	computeTimeMultTrans_
	time needed only for the computation of mult transposed, interesting on accelerator boards More...
base::DataMatrix	dataset_
	the dataset More...
double	lambda_
	the lambda, the regularisation parameter More...
base::SGppStopwatch *	myTimer_
	Stopwatch needed to determine the durations of mult and mult transposed. More...

Detailed Description

Abstract class that defines the virtual class base::OperationMatrix for classification and regression problems.

Constructor & Destructor Documentation

DMSystemMatrixBase()

sgpp::datadriven::DMSystemMatrixBase::DMSystemMatrixBase	(	base::DataMatrix &	trainData,
		double	lambda
	)

Std-Constructor.

Parameters

trainData	matrix with training data
lambda	the lambda, the regression parameter

References myTimer_.

~DMSystemMatrixBase()

sgpp::datadriven::DMSystemMatrixBase::~DMSystemMatrixBase ( )

virtual

Std-Destructor.

References myTimer_.

Member Function Documentation

generateb()

virtual void sgpp::datadriven::DMSystemMatrixBase::generateb ( base::DataVector &  classes, base::DataVector &  b  )

pure virtual

Generates the right hand side of the classification equation.

Parameters

classes	the class information of the training data
b	reference to the vector that will contain the result of the matrix vector multiplication on the rhs

Implemented in sgpp::datadriven::SystemMatrixLeastSquaresIdentity, and sgpp::datadriven::DMSystemMatrix.

Referenced by sgpp::datadriven::LearnerBase::train().

getTimers()

void sgpp::datadriven::DMSystemMatrixBase::getTimers ( double &  timeMult, double &  computeMult, double &  timeMultTrans, double &  computeMultTrans  )

virtual

gets the timer's values by saving them into call by reference values

Parameters

timeMult	variable to store overall time needed for Mult
computeMult	variable to store compute time needed for Mult
timeMultTrans	variable to store everall time needed for Mult Transposed
computeMultTrans	variable to store compute time needed for Mult Transposed

References completeTimeMult_, completeTimeMultTrans_, computeTimeMult_, and computeTimeMultTrans_.

Referenced by sgpp::datadriven::LearnerBase::train().

mult()

virtual void sgpp::datadriven::DMSystemMatrixBase::mult ( base::DataVector &  alpha, base::DataVector &  result  )

pure virtual

starts the Multiplication with the matrix

Parameters

alpha	DataVector that contains the ansatzfunctions' coefficients
result	DataVector into which the result of the Laplace operation is stored

Implements sgpp::base::OperationMatrix.

Implemented in sgpp::datadriven::SystemMatrixLeastSquaresIdentity, and sgpp::datadriven::DMSystemMatrix.

prepareGrid()

void sgpp::datadriven::DMSystemMatrixBase::prepareGrid ( )

virtual

forward declaration

rebuilds the base::DataMatrix for Level and Index this routine is needed for supporting adaptive grids with vectorized high performance kernels

Reimplemented in sgpp::datadriven::SystemMatrixLeastSquaresIdentity.

Referenced by sgpp::datadriven::LearnerBase::train().

resetTimers()

void sgpp::datadriven::DMSystemMatrixBase::resetTimers ( )

virtual

resets all timers to 0

References completeTimeMult_, completeTimeMultTrans_, computeTimeMult_, and computeTimeMultTrans_.

Member Data Documentation

completeTimeMult_

double sgpp::datadriven::DMSystemMatrixBase::completeTimeMult_

protected

time needed for Mult

Referenced by getTimers(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult(), and resetTimers().

completeTimeMultTrans_

double sgpp::datadriven::DMSystemMatrixBase::completeTimeMultTrans_

protected

time needed for Mult transposed

Referenced by sgpp::datadriven::SystemMatrixLeastSquaresIdentity::generateb(), getTimers(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult(), and resetTimers().

computeTimeMult_

double sgpp::datadriven::DMSystemMatrixBase::computeTimeMult_

protected

time needed only for the computation of mult, interesting on accelerator boards

Referenced by getTimers(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult(), and resetTimers().

computeTimeMultTrans_

double sgpp::datadriven::DMSystemMatrixBase::computeTimeMultTrans_

protected

time needed only for the computation of mult transposed, interesting on accelerator boards

Referenced by sgpp::datadriven::SystemMatrixLeastSquaresIdentity::generateb(), getTimers(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult(), and resetTimers().

dataset_

base::DataMatrix sgpp::datadriven::DMSystemMatrixBase::dataset_

protected

the dataset

Referenced by sgpp::datadriven::DMSystemMatrix::DMSystemMatrix(), sgpp::datadriven::DMSystemMatrix::generateb(), sgpp::datadriven::DMSystemMatrix::mult(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::setImplementation(), and sgpp::datadriven::SystemMatrixLeastSquaresIdentity::SystemMatrixLeastSquaresIdentity().

lambda_

double sgpp::datadriven::DMSystemMatrixBase::lambda_

protected

the lambda, the regularisation parameter

Referenced by sgpp::datadriven::DMSystemMatrix::mult(), and sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult().

myTimer_

base::SGppStopwatch* sgpp::datadriven::DMSystemMatrixBase::myTimer_

protected

Stopwatch needed to determine the durations of mult and mult transposed.

Referenced by DMSystemMatrixBase(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::generateb(), sgpp::datadriven::SystemMatrixLeastSquaresIdentity::mult(), and ~DMSystemMatrixBase().

---

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

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