#include <Multiplier.h>

Inheritance diagram for storm::solver::Multiplier< ValueType >:

[legend]

Public Member Functions
	Multiplier (storm::storage::SparseMatrix< ValueType > const &matrix)

virtual	~Multiplier ()=default

virtual void	clearCache () const

virtual void	multiply (Environment const &env, std::vector< ValueType > const &x, std::vector< ValueType > const *b, std::vector< ValueType > &result) const =0
	Performs a matrix-vector multiplication x' = A*x + b.

virtual void	multiplyGaussSeidel (Environment const &env, std::vector< ValueType > &x, std::vector< ValueType > const *b, bool backwards=true) const =0
	Performs a matrix-vector multiplication in gauss-seidel style.

void	multiplyAndReduce (Environment const &env, OptimizationDirection const &dir, std::vector< ValueType > const &x, std::vector< ValueType > const b, std::vector< ValueType > &result, std::vector< uint_fast64_t > choices=nullptr) const
	Performs a matrix-vector multiplication x' = A*x + b and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

virtual void	multiplyAndReduce (Environment const &env, OptimizationDirection const &dir, std::vector< uint64_t > const &rowGroupIndices, std::vector< ValueType > const &x, std::vector< ValueType > const b, std::vector< ValueType > &result, std::vector< uint_fast64_t > choices=nullptr) const =0

void	multiplyAndReduceGaussSeidel (Environment const &env, OptimizationDirection const &dir, std::vector< ValueType > &x, std::vector< ValueType > const b, std::vector< uint_fast64_t > choices=nullptr, bool backwards=true) const
	Performs a matrix-vector multiplication in gauss-seidel style and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

virtual void	multiplyAndReduceGaussSeidel (Environment const &env, OptimizationDirection const &dir, std::vector< uint64_t > const &rowGroupIndices, std::vector< ValueType > &x, std::vector< ValueType > const b, std::vector< uint_fast64_t > choices=nullptr, bool backwards=true) const =0

void	repeatedMultiply (Environment const &env, std::vector< ValueType > &x, std::vector< ValueType > const *b, uint64_t n) const
	Performs repeated matrix-vector multiplication, using x[0] = x and x[i + 1] = A*x[i] + b.

void	repeatedMultiplyAndReduce (Environment const &env, OptimizationDirection const &dir, std::vector< ValueType > &x, std::vector< ValueType > const *b, uint64_t n) const
	Performs repeated matrix-vector multiplication x' = A*x + b and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

virtual void	multiplyRow (uint64_t const &rowIndex, std::vector< ValueType > const &x, ValueType &value) const =0
	Multiplies the row with the given index with x and adds the result to the provided value.

virtual void	multiplyRow2 (uint64_t const &rowIndex, std::vector< ValueType > const &x1, ValueType &val1, std::vector< ValueType > const &x2, ValueType &val2) const
	Multiplies the row with the given index with x1 and x2 and adds the given offset o1 and o2, respectively.

Protected Attributes
std::unique_ptr< std::vector< ValueType > >	cachedVector

storm::storage::SparseMatrix< ValueType > const &	matrix

Detailed Description

template<typename ValueType>
class storm::solver::Multiplier< ValueType >

Definition at line 21 of file Multiplier.h.

Constructor & Destructor Documentation

◆ Multiplier()

template<typename ValueType >

storm::solver::Multiplier< ValueType >::Multiplier ( storm::storage::SparseMatrix< ValueType > const & matrix )

Definition at line 25 of file Multiplier.cpp.

◆ ~Multiplier()

template<typename ValueType >

virtual storm::solver::Multiplier< ValueType >::~Multiplier ( )

virtualdefault

Member Function Documentation

◆ clearCache()

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::clearCache ( ) const

virtual

Reimplemented in storm::solver::GmmxxMultiplier< ValueType >.

Definition at line 30 of file Multiplier.cpp.

◆ multiply()

template<typename ValueType >

virtual void storm::solver::Multiplier< ValueType >::multiply	(	Environment const &	env,
		std::vector< ValueType > const &	x,
		std::vector< ValueType > const *	b,
		std::vector< ValueType > &	result
	)		const

pure virtual

Performs a matrix-vector multiplication x' = A*x + b.

Parameters

x	The input vector with which to multiply the matrix. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after the multiplication. If given, its length must be equal to the number of rows of A.
result	The target vector into which to write the multiplication result. Its length must be equal to the number of rows of A. Can be the same as the x vector.

Implemented in storm::solver::GmmxxMultiplier< ValueType >, and storm::solver::NativeMultiplier< ValueType >.

◆ multiplyAndReduce() [1/2]

template<typename ValueType >

virtual void storm::solver::Multiplier< ValueType >::multiplyAndReduce	(	Environment const &	env,
		OptimizationDirection const &	dir,
		std::vector< uint64_t > const &	rowGroupIndices,
		std::vector< ValueType > const &	x,
		std::vector< ValueType > const *	b,
		std::vector< ValueType > &	result,
		std::vector< uint_fast64_t > *	choices = `nullptr`
	)		const

pure virtual

Implemented in storm::solver::GmmxxMultiplier< ValueType >, and storm::solver::NativeMultiplier< ValueType >.

◆ multiplyAndReduce() [2/2]

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::multiplyAndReduce	(	Environment const &	env,
		OptimizationDirection const &	dir,
		std::vector< ValueType > const &	x,
		std::vector< ValueType > const *	b,
		std::vector< ValueType > &	result,
		std::vector< uint_fast64_t > *	choices = `nullptr`
	)		const

Performs a matrix-vector multiplication x' = A*x + b and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

Parameters

dir	The direction for the reduction step.
rowGroupIndices	A vector storing the row groups over which to reduce.
x	The input vector with which to multiply the matrix. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after the multiplication. If given, its length must be equal to the number of rows of A.
result	The target vector into which to write the multiplication result. Its length must be equal to the number of rows of A. Can be the same as the x vector.
choices	If given, the choices made in the reduction process are written to this vector.

Definition at line 35 of file Multiplier.cpp.

◆ multiplyAndReduceGaussSeidel() [1/2]

template<typename ValueType >

virtual void storm::solver::Multiplier< ValueType >::multiplyAndReduceGaussSeidel	(	Environment const &	env,
		OptimizationDirection const &	dir,
		std::vector< uint64_t > const &	rowGroupIndices,
		std::vector< ValueType > &	x,
		std::vector< ValueType > const *	b,
		std::vector< uint_fast64_t > *	choices = `nullptr`,
		bool	backwards = `true`
	)		const

pure virtual

Implemented in storm::solver::GmmxxMultiplier< ValueType >, and storm::solver::NativeMultiplier< ValueType >.

◆ multiplyAndReduceGaussSeidel() [2/2]

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::multiplyAndReduceGaussSeidel	(	Environment const &	env,
		OptimizationDirection const &	dir,
		std::vector< ValueType > &	x,
		std::vector< ValueType > const *	b,
		std::vector< uint_fast64_t > *	choices = `nullptr`,
		bool	backwards = `true`
	)		const

Performs a matrix-vector multiplication in gauss-seidel style and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

Parameters

dir	The direction for the reduction step.
rowGroupIndices	A vector storing the row groups over which to reduce.
x	The input/output vector with which to multiply the matrix. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after the multiplication. If given, its length must be equal to the number of rows of A.
result	The target vector into which to write the multiplication result. Its length must be equal to the number of rows of A. Can be the same as the x vector.
choices	If given, the choices made in the reduction process are written to this vector.
backwards	if true, the iterations will be performed beginning from the last rowgroup and ending at the first rowgroup.

Definition at line 41 of file Multiplier.cpp.

◆ multiplyGaussSeidel()

template<typename ValueType >

virtual void storm::solver::Multiplier< ValueType >::multiplyGaussSeidel	(	Environment const &	env,
		std::vector< ValueType > &	x,
		std::vector< ValueType > const *	b,
		bool	backwards = `true`
	)		const

pure virtual

Performs a matrix-vector multiplication in gauss-seidel style.

Parameters

x	The input/output vector with which to multiply the matrix. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after the multiplication. If given, its length must be equal to the number of rows of A.
backwards	if true, the iterations will be performed beginning from the last row and ending at the first row.

Implemented in storm::solver::GmmxxMultiplier< ValueType >, and storm::solver::NativeMultiplier< ValueType >.

◆ multiplyRow()

template<typename ValueType >

virtual void storm::solver::Multiplier< ValueType >::multiplyRow	(	uint64_t const &	rowIndex,
		std::vector< ValueType > const &	x,
		ValueType &	value
	)		const

pure virtual

Multiplies the row with the given index with x and adds the result to the provided value.

Parameters

rowIndex	The index of the considered row
x	The input vector with which the row is multiplied
value	The multiplication result is added to this value. It shall not reffer to a value in x or in the Matrix.

Implemented in storm::solver::GmmxxMultiplier< ValueType >, and storm::solver::NativeMultiplier< ValueType >.

◆ multiplyRow2()

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::multiplyRow2	(	uint64_t const &	rowIndex,
		std::vector< ValueType > const &	x1,
		ValueType &	val1,
		std::vector< ValueType > const &	x2,
		ValueType &	val2
	)		const

virtual

Multiplies the row with the given index with x1 and x2 and adds the given offset o1 and o2, respectively.

Parameters

rowIndex	The index of the considered row
x1	The first input vector with which the row is multiplied.
val1	The first multiplication result is added to this value. It shall not reffer to a value in x or in the Matrix.
x2	The second input vector with which the row is multiplied.
val2	The second multiplication result is added to this value. It shall not reffer to a value in x or in the Matrix.

Reimplemented in storm::solver::NativeMultiplier< ValueType >.

Definition at line 77 of file Multiplier.cpp.

◆ repeatedMultiply()

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::repeatedMultiply	(	Environment const &	env,
		std::vector< ValueType > &	x,
		std::vector< ValueType > const *	b,
		uint64_t	n
	)		const

Performs repeated matrix-vector multiplication, using x[0] = x and x[i + 1] = A*x[i] + b.

After performing the necessary multiplications, the result is written to the input vector x. Note that the matrix A has to be given upon construction time of the solver object.

Parameters

x	The initial vector with which to perform matrix-vector multiplication. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after each multiplication. If given, its length must be equal to the number of rows of A.
n	The number of times to perform the multiplication.

Definition at line 47 of file Multiplier.cpp.

◆ repeatedMultiplyAndReduce()

template<typename ValueType >

void storm::solver::Multiplier< ValueType >::repeatedMultiplyAndReduce	(	Environment const &	env,
		OptimizationDirection const &	dir,
		std::vector< ValueType > &	x,
		std::vector< ValueType > const *	b,
		uint64_t	n
	)		const

Performs repeated matrix-vector multiplication x' = A*x + b and then minimizes/maximizes over the row groups so that the resulting vector has the size of number of row groups of A.

Parameters

dir	The direction for the reduction step.
x	The input vector with which to multiply the matrix. Its length must be equal to the number of columns of A.
b	If non-null, this vector is added after the multiplication. If given, its length must be equal to the number of rows of A.
result	The target vector into which to write the multiplication result. Its length must be equal to the number of rows of A.
n	The number of times to perform the multiplication.

Definition at line 62 of file Multiplier.cpp.

Member Data Documentation

◆ cachedVector

template<typename ValueType >

std::unique_ptr<std::vector<ValueType> > storm::solver::Multiplier< ValueType >::cachedVector

mutableprotected

Definition at line 145 of file Multiplier.h.

◆ matrix

template<typename ValueType >

storm::storage::SparseMatrix<ValueType> const& storm::solver::Multiplier< ValueType >::matrix

protected

Definition at line 146 of file Multiplier.h.

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

src/storm/solver/multiplier/Multiplier.h
src/storm/solver/multiplier/Multiplier.cpp

Public Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ Multiplier()

◆ ~Multiplier()

Member Function Documentation

◆ clearCache()

◆ multiply()

◆ multiplyAndReduce() [1/2]

◆ multiplyAndReduce() [2/2]

◆ multiplyAndReduceGaussSeidel() [1/2]

◆ multiplyAndReduceGaussSeidel() [2/2]

◆ multiplyGaussSeidel()

◆ multiplyRow()

◆ multiplyRow2()

◆ repeatedMultiply()

◆ repeatedMultiplyAndReduce()

Member Data Documentation

◆ cachedVector

◆ matrix