CDPL::Math::MinimizerVariableArrayTraits< A > Struct Template Reference

Traits template that adapts arbitrary variable-array types to the linear-algebra operations required by minimizer implementations (dot, norm2, axpy, clear, assign, multiply, sub). More...

#include <MinimizerVariableArrayTraits.hpp>

Inheritance diagram for CDPL::Math::MinimizerVariableArrayTraits< A >:

Public Types
typedef A	ArrayType
	The variable-array type. More...
typedef A::ValueType	ValueType
	The scalar value type stored in the array. More...
typedef A::SizeType	SizeType
	The size type used by the array. More...

Static Public Member Functions
template<typename T >
static T	dot (const ArrayType &a1, const ArrayType &a2)
	Computes the inner product (dot product) of two variable arrays. More...
template<typename T >
static T	norm2 (const ArrayType &a)
	Computes the Euclidean (L2) norm of a using a numerically stable scaling algorithm. More...
template<typename T >
static void	axpy (const T &alpha, const ArrayType &x, ArrayType &y)
	Performs the in-place BLAS-style axpy operation \( y \leftarrow y + \alpha\, x \). More...
static void	clear (ArrayType &a)
	Sets all elements of a to the default-constructed ValueType. More...
static void	assign (ArrayType &a1, const ArrayType &a2)
	Copies the contents of a2 into a1. More...
template<typename T >
static void	multiply (ArrayType &a, const T &v)
	Multiplies every element of a by the scalar v. More...
static void	sub (ArrayType &a1, const ArrayType &a2)
	Subtracts a2 from a1 element-wise ( \( a_1 \leftarrow a_1 - a_2 \)). More...

Detailed Description

template<typename A>
struct CDPL::Math::MinimizerVariableArrayTraits< A >

Traits template that adapts arbitrary variable-array types to the linear-algebra operations required by minimizer implementations (dot, norm2, axpy, clear, assign, multiply, sub).

The primary template assumes A behaves like a Math::Vector (provides getSize, clear, assign, plusAssign/minusAssign, innerProd and arithmetic operators). Specializations adapt other array-of-vector storage types.

Template Parameters

A	The variable array type.

Member Typedef Documentation

ArrayType

template<typename A >

typedef A CDPL::Math::MinimizerVariableArrayTraits< A >::ArrayType

The variable-array type.

ValueType

template<typename A >

typedef A::ValueType CDPL::Math::MinimizerVariableArrayTraits< A >::ValueType

The scalar value type stored in the array.

SizeType

template<typename A >

typedef A::SizeType CDPL::Math::MinimizerVariableArrayTraits< A >::SizeType

The size type used by the array.

Member Function Documentation

dot()

template<typename A >

template<typename T >

static T CDPL::Math::MinimizerVariableArrayTraits< A >::dot ( const ArrayType &  a1, const ArrayType &  a2  )

inlinestatic

Computes the inner product (dot product) of two variable arrays.

Template Parameters

T	The scalar result type.

Parameters

a1	The first variable array.
a2	The second variable array.

Returns

The inner product as a value of type T.

norm2()

template<typename A >

template<typename T >

static T CDPL::Math::MinimizerVariableArrayTraits< A >::norm2 ( const ArrayType &  a )

inlinestatic

Computes the Euclidean (L2) norm of a using a numerically stable scaling algorithm.

Template Parameters

T	The scalar result type.

Parameters

a	The variable array.

Returns

The Euclidean norm as a value of type T.

axpy()

template<typename A >

template<typename T >

static void CDPL::Math::MinimizerVariableArrayTraits< A >::axpy ( const T &  alpha, const ArrayType &  x, ArrayType &  y  )

inlinestatic

Performs the in-place BLAS-style axpy operation \( y \leftarrow y + \alpha\, x \).

Template Parameters

T	The scalar type of alpha.

Parameters

alpha	The scalar multiplier.
x	The variable array x.
y	The variable array y (updated in place).

clear()

template<typename A >

static void CDPL::Math::MinimizerVariableArrayTraits< A >::clear ( ArrayType &  a )

inlinestatic

Sets all elements of a to the default-constructed ValueType.

Parameters

a	The variable array to clear.

assign()

template<typename A >

static void CDPL::Math::MinimizerVariableArrayTraits< A >::assign ( ArrayType &  a1, const ArrayType &  a2  )

inlinestatic

Copies the contents of a2 into a1.

Parameters

a1	The destination variable array.
a2	The source variable array.

multiply()

template<typename A >

template<typename T >

static void CDPL::Math::MinimizerVariableArrayTraits< A >::multiply ( ArrayType &  a, const T &  v  )

inlinestatic

Multiplies every element of a by the scalar v.

Template Parameters

T	The scalar type of v.

Parameters

a	The variable array to scale (updated in place).
v	The scalar multiplier.

sub()

template<typename A >

static void CDPL::Math::MinimizerVariableArrayTraits< A >::sub ( ArrayType &  a1, const ArrayType &  a2  )

inlinestatic

Subtracts a2 from a1 element-wise ( \( a_1 \leftarrow a_1 - a_2 \)).

Parameters

a1	The destination variable array (updated in place).
a2	The variable array to subtract.

---

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

• MinimizerVariableArrayTraits.hpp