sgpp::datadriven::OperationDensityConditionalLinear Class Reference

Marginalize Probability Density Function. More...

#include <OperationDensityConditionalLinear.hpp>

Inheritance diagram for sgpp::datadriven::OperationDensityConditionalLinear:

Public Member Functions
void	doConditional (base::DataVector &alpha, base::Grid *&mg, base::DataVector &malpha, unsigned int mdim, double xbar)
	Marginalizes (Density) Functions. More...
	OperationDensityConditionalLinear (base::Grid *grid)
virtual	~OperationDensityConditionalLinear ()
Public Member Functions inherited from sgpp::datadriven::OperationDensityConditional
	OperationDensityConditional ()
virtual	~OperationDensityConditional ()

Protected Attributes
base::Grid *	grid

Detailed Description

Marginalize Probability Density Function.

Constructor & Destructor Documentation

sgpp::datadriven::OperationDensityConditionalLinear::OperationDensityConditionalLinear ( base::Grid *  grid )

inlineexplicit

virtual sgpp::datadriven::OperationDensityConditionalLinear::~OperationDensityConditionalLinear ( )

inlinevirtual

References alpha, and doConditional().

Member Function Documentation

void sgpp::datadriven::OperationDensityConditionalLinear::doConditional ( base::DataVector &  alpha, base::Grid *&  mg, base::DataVector &  malpha, unsigned int  mdim, double  xbar  )

virtual

Marginalizes (Density) Functions.

Parameters

alpha	Coefficient vector for current grid
mg	Referenz of grid pointer
malpha	Coefficient vector for new grid (mg). Will be resized.
mdim	Marginalize in dimension mdim
xbar	Point at which to conditionalize

Assume: mdim = 1 Compute vector with values (phi_{l1,i1}(xbar) = phi_{l1,i1}(xbar)*phi_{l2,i2}(0.5)

Compute theta = theta + alpha_{l,i}*zeta_{l,i}*int{phi_{l2, i_2}}

Generate d - 1 dimensional grid, as in marginalize

Note: Because of adaptively refined sparse grids, we cannot simply generate a regular grid. Thus, we need to add point after point to the new grid mg

Compute coefficients malpha for grid mg

Implements sgpp::datadriven::OperationDensityConditional.

References sgpp::base::Grid::createLinearGrid(), sgpp::base::HashGridStorage::getDimension(), sgpp::base::HashGridPoint::getIndex(), sgpp::base::HashGridPoint::getLevel(), sgpp::base::HashGridStorage::getPoint(), sgpp::base::HashGridStorage::getSize(), sgpp::base::DataVector::getSize(), sgpp::base::Grid::getStorage(), grid, sgpp::base::DataVector::mult(), sgpp::combigrid::pow(), sgpp::base::HashGridPoint::set(), and sgpp::base::DataVector::setAll().

Referenced by ~OperationDensityConditionalLinear().

Member Data Documentation

base::Grid* sgpp::datadriven::OperationDensityConditionalLinear::grid

protected

Referenced by doConditional(), python.uq.learner.Interpolant.Interpolant::doLearningIteration(), python.learner.Classifier.Classifier::evalError(), python.uq.learner.Interpolant.Interpolant::evalError(), python.uq.learner.SimulationLearner.SimulationLearner::getGrid(), python.uq.learner.Regressor.Regressor::learnData(), python.uq.learner.Regressor.Regressor::learnDataWithFolding(), and python.uq.learner.Regressor.Regressor::learnDataWithTest().

---

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

• datadriven/src/sgpp/datadriven/operation/hash/simple/OperationDensityConditionalLinear.hpp
• datadriven/src/sgpp/datadriven/operation/hash/simple/OperationDensityConditionalLinear.cpp