sgpp::base::OperationMatrixSP Class Referenceabstract

Abstract definition of a matrix operator interface. More...

#include <OperationMatrixSP.hpp>

Inheritance diagram for sgpp::base::OperationMatrixSP:

Public Member Functions

virtual void mult (DataVectorSP &alpha, DataVectorSP &result)=0
 starts the Multiplication with the Laplace matrix More...
 OperationMatrixSP ()
 Constructor. More...
virtual ~OperationMatrixSP ()
 Destructor. More...

Detailed Description

Abstract definition of a matrix operator interface.

Everytime you need to apply a matrix to the ansatzfunction's coefficients in single precision derive a class from OperationMatrixSP.

This is an re-implementation of the standard OperationMatrix for single precision floating point numbers in order to increase support for GPUs.

Constructor & Destructor Documentation

sgpp::base::OperationMatrixSP::OperationMatrixSP ( )


virtual sgpp::base::OperationMatrixSP::~OperationMatrixSP ( )


References alpha, and mult().

Member Function Documentation

virtual void sgpp::base::OperationMatrixSP::mult ( DataVectorSP alpha,
DataVectorSP result 
pure virtual

starts the Multiplication with the Laplace matrix

alphaDataVectorSP that contains the ansatzfunctions' coefficients
resultDataVectorSP into which the result of the Laplace operation is stored

Implemented in sgpp::datadriven::DMSystemMatrixBaseSP.

Referenced by sgpp::solver::BiCGStabSP::solve(), sgpp::solver::ConjugateGradientsSP::solve(), and ~OperationMatrixSP().

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