sgpp::optimization::ComponentScalarFunction Class Reference

One component of a vector-valued function. More...

#include <ComponentScalarFunction.hpp>

Inheritance diagram for sgpp::optimization::ComponentScalarFunction:

Public Member Functions
void	clone (std::unique_ptr< ScalarFunction > &clone) const override
	ComponentScalarFunction (ScalarFunction &f, std::vector< double > defaultValues=std::vector< double >())
	Constructor. More...
	ComponentScalarFunction (VectorFunction &f, size_t k, std::vector< double > defaultValues=std::vector< double >())
	Constructor. More...
double	eval (const base::DataVector &x) override
	~ComponentScalarFunction () override
	Destructor. More...
Public Member Functions inherited from sgpp::optimization::ScalarFunction
size_t	getNumberOfParameters () const
	ScalarFunction (size_t d)
	Constructor. More...
virtual	~ScalarFunction ()
	Destructor. More...

Protected Member Functions
void	initialize ()

Protected Attributes
std::vector< double >	defaultValues
	vector of default values, indicating free variables with NAN More...
size_t	dF
	dimension of underlying function More...
ScalarFunction *	fScalar
	scalar-valued function More...
VectorFunction *	fVector
	vector-valued function More...
size_t	k
	index of component More...
base::DataVector	tmpVec1
	temporary vector 1 More...
base::DataVector	tmpVec2
	temporary vector 2 More...
Protected Attributes inherited from sgpp::optimization::ScalarFunction
size_t	d
	dimension of the domain More...

Detailed Description

One component of a vector-valued function.

Given a vector-valued \(f\colon [0, 1]^d \to \mathbb{R}^m\) and indices \(1 \le i_1 < \dotsb < i_n \le d\), this class represents a new function \(g\colon [0, 1]^n \to \mathbb{R}\) with \(g(y_{i_1}, \dotsc, y_{i_n}) := f_k(y_1, \dotsc, y_d)\), where \(y_t\) is constant for \(t \in \{i_1, \dotsc, i_n\}\).

The resulting function \(g\) is similar to a "slice plot" of the component \(f_k\) of \(f\).

Constructor & Destructor Documentation

ComponentScalarFunction() [1/2]

sgpp::optimization::ComponentScalarFunction::ComponentScalarFunction ( ScalarFunction &  f, std::vector< double >  defaultValues = std::vector<double>()  )

inline

Constructor.

Use it like this: ComponentScalarFunction g(f, {NAN, NAN, 0.42}); where f is a scalar-valued function with 3 parameters. This selects the first two parameters of f, while constantly using 0.42 for the third parameter.

Parameters

f	scalar-valued function
defaultValues	Vector of constant default values. It can be either empty (the default) or a vector of exactly m doubles, each of which can be finite or NAN. If the vector is empty, it will be initialized as m NANs (i.e., no restriction of the parameter domain). Each NAN represents a free parameter \(x_t\), while the finite entries denote the constant values for the corresponding parameter.

References initialize().

Referenced by clone().

ComponentScalarFunction() [2/2]

sgpp::optimization::ComponentScalarFunction::ComponentScalarFunction ( VectorFunction &  f, size_t  k, std::vector< double >  defaultValues = std::vector<double>()  )

inline

Constructor.

Use it like this: ComponentScalarFunction g(f, 3, {NAN, NAN, 0.42}); where f is a vector-valued function with 5 components and 3 parameters. This selects the first two parameters and the fourth component of f, while constantly using 0.42 for the third parameter.

Parameters

f	vector-valued function (m components)
k	index of component \(f_k\) to select (between 0 and m - 1)
defaultValues	see other constructor

References initialize().

~ComponentScalarFunction()

sgpp::optimization::ComponentScalarFunction::~ComponentScalarFunction ( )

inlineoverride

Destructor.

Member Function Documentation

clone()

void sgpp::optimization::ComponentScalarFunction::clone ( std::unique_ptr< ScalarFunction > &  clone ) const

inlineoverridevirtual

Parameters

[out]

clone

pointer to cloned object

Implements sgpp::optimization::ScalarFunction.

References ComponentScalarFunction().

eval()

double sgpp::optimization::ComponentScalarFunction::eval ( const base::DataVector &  x )

inlineoverridevirtual

Parameters

x	evaluation point \(\vec{x} \in [0, 1]^n\)

Returns

\(g(\vec{x}) := f_k(y_1, \dotsc, y_d)\) where \((x_1, \dotsc, x_n) = (y_{i_1}, \dotsc, y_{i_n})\)

Implements sgpp::optimization::ScalarFunction.

References defaultValues, dF, sgpp::optimization::ScalarFunction::eval(), sgpp::optimization::VectorFunction::eval(), fScalar, fVector, k, tmpVec1, and tmpVec2.

initialize()

void sgpp::optimization::ComponentScalarFunction::initialize ( )

inlineprotected

References dF.

Referenced by ComponentScalarFunction().

Member Data Documentation

defaultValues

std::vector<double> sgpp::optimization::ComponentScalarFunction::defaultValues

protected

vector of default values, indicating free variables with NAN

Referenced by eval().

dF

size_t sgpp::optimization::ComponentScalarFunction::dF

protected

dimension of underlying function

Referenced by eval(), and initialize().

fScalar

ScalarFunction* sgpp::optimization::ComponentScalarFunction::fScalar

protected

scalar-valued function

Referenced by eval().

fVector

VectorFunction* sgpp::optimization::ComponentScalarFunction::fVector

protected

vector-valued function

Referenced by eval().

k

size_t sgpp::optimization::ComponentScalarFunction::k

protected

index of component

Referenced by eval().

tmpVec1

base::DataVector sgpp::optimization::ComponentScalarFunction::tmpVec1

protected

temporary vector 1

Referenced by eval().

tmpVec2

base::DataVector sgpp::optimization::ComponentScalarFunction::tmpVec2

protected

temporary vector 2

Referenced by eval().

---

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

• optimization/src/sgpp/optimization/function/scalar/ComponentScalarFunction.hpp