SG++-Doxygen-Documentation
sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T > Class Template Reference

#include <OperationMultiEvalStreamingModOCLFastMultiPlatform.hpp>

Inheritance diagram for sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >:
sgpp::base::OperationMultipleEval

Public Member Functions

double getDuration () override
 
void mult (base::DataVector &alpha, base::DataVector &result) override
 Multiplication of \(B^T\) with vector \(\alpha\). More...
 
void multTranspose (base::DataVector &source, base::DataVector &result) override
 Multiplication of \(B\) with vector \(\alpha\). More...
 
 OperationMultiEvalStreamingModOCLFastMultiPlatform (base::Grid &grid, base::DataMatrix &dataset, std::shared_ptr< base::OCLManagerMultiPlatform > manager, std::shared_ptr< base::OCLOperationConfiguration > parameters)
 
void prepare () override
 Used for kernel-specific setup like special data structures that are defined from the current state of the grid. More...
 
 ~OperationMultiEvalStreamingModOCLFastMultiPlatform ()
 
- Public Member Functions inherited from sgpp::base::OperationMultipleEval
void eval (DataVector &alpha, DataVector &result)
 Evaluate multiple datapoints with the specified grid. More...
 
virtual std::string getImplementationName ()
 Name of this implementation of the operation. More...
 
virtual void mult (DataVector &alpha, DataVector &result, size_t startIndexData, size_t endIndexData)
 Multiplication of \(B^T\) with vector \(\alpha\). More...
 
virtual void multTranspose (DataVector &source, DataVector &result, size_t startIndexGrid, size_t endIndexGrid)
 Multiplication of \(B\) with vector \(\alpha\). More...
 
 OperationMultipleEval (sgpp::base::Grid &grid, DataMatrix &dataset)
 Constructor. More...
 
virtual ~OperationMultipleEval ()
 Destructor. More...
 

Protected Attributes

size_t commonDatasetPadding
 
size_t commonGridPadding
 
size_t datasetSizePadded
 
size_t datasetSizeUnpadded
 
std::vector< std::shared_ptr< base::OCLDevice > > devices
 
size_t dims
 
double duration
 
size_t gridSizePadded
 
size_t gridSizeUnpadded
 
std::vector< T > index
 Member to store the sparse grid's indices for better vectorization. More...
 
std::vector< T > kernelDataset
 
std::vector< T > level
 Member to store the sparse grid's levels for better vectorization. More...
 
std::shared_ptr< base::OCLManagerMultiPlatformmanager
 
std::vector< StreamingModOCLFastMultiPlatform::KernelMult< T > > multKernels
 
std::vector< StreamingModOCLFastMultiPlatform::KernelMultTranspose< T > > multTransposeKernels
 
sgpp::base::SGppStopwatch myTimer
 Timer object to handle time measurements. More...
 
std::shared_ptr< base::OCLOperationConfigurationparameters
 
sgpp::base::DataMatrix preparedDataset
 
std::shared_ptr< base::QueueLoadBalancerMutexqueueLoadBalancerMult
 
std::shared_ptr< base::QueueLoadBalancerMutexqueueLoadBalancerMultTranspose
 
bool verbose
 
- Protected Attributes inherited from sgpp::base::OperationMultipleEval
DataMatrixdataset
 
Gridgrid
 
bool isPrepared
 

Constructor & Destructor Documentation

◆ OperationMultiEvalStreamingModOCLFastMultiPlatform()

◆ ~OperationMultiEvalStreamingModOCLFastMultiPlatform()

Member Function Documentation

◆ getDuration()

◆ mult()

template<typename T>
void sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::mult ( base::DataVector alpha,
base::DataVector result 
)
inlineoverridevirtual

Multiplication of \(B^T\) with vector \(\alpha\).

Parameters
alphavector, to which \(B\) is applied. Typically the coefficient vector
resultthe result vector of the matrix vector multiplication

Implements sgpp::base::OperationMultipleEval.

References sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::datasetSizePadded, sgpp::base::DataVector::getSize(), sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::gridSizePadded, python.statsfileInfo::i, sgpp::base::SGppStopwatch::start(), python.leja::start, and sgpp::base::SGppStopwatch::stop().

◆ multTranspose()

◆ prepare()

template<typename T>
void sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::prepare ( )
inlineoverridevirtual

Used for kernel-specific setup like special data structures that are defined from the current state of the grid.

This function is by default called with each "mult()", "multTranspose()" or evaluation operation and can be ignored from an external perspective. This is not overridden by every kernel.

Reimplemented from sgpp::base::OperationMultipleEval.

References sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::commonDatasetPadding, sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::commonGridPadding, sgpp::base::OperationMultipleEval::dataset, sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::datasetSizePadded, chess::dim, sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::dims, sgpp::base::HashGridPoint::get(), sgpp::base::HashGridStorage::getDimension(), sgpp::datadriven::StreamingModOCLFastMultiPlatform::Configuration::getKernelName(), sgpp::base::DataMatrix::getNrows(), sgpp::base::HashGridStorage::getPoint(), sgpp::base::DataMatrix::getRow(), sgpp::base::HashGridStorage::getSize(), sgpp::base::Grid::getStorage(), sgpp::base::OperationMultipleEval::grid, sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::gridSizePadded, sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::gridSizeUnpadded, python.statsfileInfo::i, sgpp::base::DataMatrix::resize(), and sgpp::base::DataMatrix::setRow().

Referenced by sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::multTranspose(), and sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::OperationMultiEvalStreamingModOCLFastMultiPlatform().

Member Data Documentation

◆ commonDatasetPadding

◆ commonGridPadding

◆ datasetSizePadded

◆ datasetSizeUnpadded

template<typename T>
size_t sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::datasetSizeUnpadded
protected

◆ devices

template<typename T>
std::vector<std::shared_ptr<base::OCLDevice> > sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::devices
protected

◆ dims

◆ duration

◆ gridSizePadded

◆ gridSizeUnpadded

◆ index

template<typename T>
std::vector<T> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::index
protected

Member to store the sparse grid's indices for better vectorization.

◆ kernelDataset

template<typename T>
std::vector<T> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::kernelDataset
protected

◆ level

template<typename T>
std::vector<T> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::level
protected

Member to store the sparse grid's levels for better vectorization.

◆ manager

template<typename T>
std::shared_ptr<base::OCLManagerMultiPlatform> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::manager
protected

◆ multKernels

◆ multTransposeKernels

◆ myTimer

Timer object to handle time measurements.

◆ parameters

template<typename T>
std::shared_ptr<base::OCLOperationConfiguration> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::parameters
protected

◆ preparedDataset

◆ queueLoadBalancerMult

template<typename T>
std::shared_ptr<base::QueueLoadBalancerMutex> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::queueLoadBalancerMult
protected

◆ queueLoadBalancerMultTranspose

template<typename T>
std::shared_ptr<base::QueueLoadBalancerMutex> sgpp::datadriven::OperationMultiEvalStreamingModOCLFastMultiPlatform< T >::queueLoadBalancerMultTranspose
protected

◆ verbose


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