Classes
struct	FeasibilityConstraints

struct	SeparatedFeasibilityConstraints

struct	LeastSquareProblem

struct	QPProblem

struct	scalar_normal_random_op

struct	ProblemCharacteristics

struct	RandomLeastSquare

Functions
LeastSquareProblem JRLQP_DLLAPI	generateBoxAndSingleConstraintProblem (int nbVar, bool act, double actLevel=0.5)

void JRLQP_DLLAPI	checkDimensions (const MatrixConstRef &G, const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false)

void JRLQP_DLLAPI	checkDimensions (const VectorConstRef &x, const VectorConstRef &u, const MatrixConstRef &G, const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false)

void JRLQP_DLLAPI	checkDimensions (const VectorConstRef &x, const VectorConstRef &u, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false)

bool JRLQP_DLLAPI	testKKT (const VectorConstRef &x, const VectorConstRef &u, const MatrixConstRef &G, const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false, double tau_p=1e-6, double tau_d=1e-6)

bool JRLQP_DLLAPI	testKKT (const VectorConstRef &x, const VectorConstRef &u, const QPProblem<> &pb, double tau_p=1e-6, double tau_d=1e-6)

bool JRLQP_DLLAPI	testKKTStationarity (const VectorConstRef &x, const VectorConstRef &u, const MatrixConstRef &G, const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false, double tau_d=1e-6)

bool JRLQP_DLLAPI	testKKTStationarity (const VectorConstRef &x, const VectorConstRef &u, const QPProblem<> &pb, double tau_d=1e-6)

bool JRLQP_DLLAPI	testKKTFeasibility (const VectorConstRef &x, const VectorConstRef &u, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC=false, double tau_p=1e-6, double tau_d=1e-6)

bool JRLQP_DLLAPI	testKKTFeasibility (const VectorConstRef &x, const VectorConstRef &u, const FeasibilityConstraints &cstr, double tau_p=1e-6, double tau_d=1e-6)

auto	randnVec (Eigen::Index size, double mean=0, double stddev=1)

auto	randnMat (Eigen::Index rows, Eigen::Index cols, double mean=0, double stddev=1)

Eigen::VectorXd	randUnitVec (Eigen::Index size)

Eigen::MatrixXd	randOrtho (Eigen::Index size, bool special=false)

Eigen::MatrixXd	randn (Eigen::Index rows, Eigen::Index cols, Eigen::Index rank=-1)

std::pair< Eigen::MatrixXd, Eigen::MatrixXd >	randDependent (Eigen::Index cols, Eigen::Index rowsA, Eigen::Index rankA, Eigen::Index rowsB, Eigen::Index rankB, Eigen::Index rankAB)

RandomLeastSquare JRLQP_DLLAPI	randomProblem (const ProblemCharacteristics &characs)

void	checkDimensions ([[maybe_unused]] int n, [[maybe_unused]] const MatrixConstRef &C, [[maybe_unused]] const VectorConstRef &bl, [[maybe_unused]] const VectorConstRef &bu, [[maybe_unused]] const VectorConstRef &xl, [[maybe_unused]] const VectorConstRef &xu, [[maybe_unused]] bool transposedC)

void	checkDimensions ([[maybe_unused]] const MatrixConstRef &G, [[maybe_unused]] const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC)

void	checkDimensions ([[maybe_unused]] const VectorConstRef &x, [[maybe_unused]] const VectorConstRef &u, const MatrixConstRef &G, const VectorConstRef &a, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC)

void	checkDimensions ([[maybe_unused]] const VectorConstRef &x, [[maybe_unused]] const VectorConstRef &u, const MatrixConstRef &C, const VectorConstRef &bl, const VectorConstRef &bu, const VectorConstRef &xl, const VectorConstRef &xu, bool transposedC)

template<typename Derived >
void	disp (const std::string &name, const MatrixBase< Derived > &M)

Variables
struct JRLQP_DLLAPI	SeparatedFeasibilityConstraints

Function Documentation

◆ checkDimensions() [1/7]

void jrl::qp::test::checkDimensions	(	[[maybe_unused] ] const MatrixConstRef &	G,
		[[maybe_unused] ] const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC
	)

◆ checkDimensions() [2/7]

void jrl::qp::test::checkDimensions	(	[[maybe_unused] ] const VectorConstRef &	x,
		[[maybe_unused] ] const VectorConstRef &	u,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC
	)

◆ checkDimensions() [3/7]

void jrl::qp::test::checkDimensions	(	[[maybe_unused] ] const VectorConstRef &	x,
		[[maybe_unused] ] const VectorConstRef &	u,
		const MatrixConstRef &	G,
		const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC
	)

◆ checkDimensions() [4/7]

void jrl::qp::test::checkDimensions	(	[[maybe_unused] ] int	n,
		[[maybe_unused] ] const MatrixConstRef &	C,
		[[maybe_unused] ] const VectorConstRef &	bl,
		[[maybe_unused] ] const VectorConstRef &	bu,
		[[maybe_unused] ] const VectorConstRef &	xl,
		[[maybe_unused] ] const VectorConstRef &	xu,
		[[maybe_unused] ] bool	transposedC
	)

◆ checkDimensions() [5/7]

void JRLQP_DLLAPI jrl::qp::test::checkDimensions	(	const MatrixConstRef &	G,
		const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`
	)

Check that all the matrix and vector dimensions of the following QP problem are consistent: min. .5 x^T G x + a^T x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. When transposedC is true, the constraints read bl <= C^T x <= bu

◆ checkDimensions() [6/7]

void JRLQP_DLLAPI jrl::qp::test::checkDimensions	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`
	)

Check that all the matrix and vector dimensions of the following FP problem are consistent: find x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. Further check that the solution vector x and the vector of Lagrange multipliers have the correct dimensions. When transposedC is true, the constraints read bl <= C^T x <= bu

◆ checkDimensions() [7/7]

void JRLQP_DLLAPI jrl::qp::test::checkDimensions	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const MatrixConstRef &	G,
		const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`
	)

Check that all the matrix and vector dimensions of the following QP problem are consistent: min. .5 x^T G x + a^T x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. Further check that the solution vector x and the vector of Lagrange multipliers have the correct dimensions. When transposedC is true, the constraints read bl <= C^T x <= bu

◆ disp()

template<typename Derived >

void jrl::qp::test::disp	(	const std::string &	name,
		const MatrixBase< Derived > &	M
	)

◆ generateBoxAndSingleConstraintProblem()

LeastSquareProblem jrl::qp::test::generateBoxAndSingleConstraintProblem	(	int	nbVar,
		bool	act,
		double	actLevel = `0.5`
	)

Generate a problem min. || x- x0 || s.t. c'x >= bl (1) xl<=x<=xu with nbVar variables and the constraint c'x >= bl active if act is true Noting xb the solution without the constraint (1) and su the vertex of the box [xl, xu] the furthest away in the direction given by c, actLevel indicates where the constraint plan intersects the segement [xb, su].

◆ randDependent()

std::pair<Eigen::MatrixXd, Eigen::MatrixXd> jrl::qp::test::randDependent	(	Eigen::Index	cols,
		Eigen::Index	rowsA,
		Eigen::Index	rankA,
		Eigen::Index	rowsB,
		Eigen::Index	rankB,
		Eigen::Index	rankAB
	)

inline

Returns two matrices A and B with specified sizes and ranks and such that [A;B] has a given rank

Parameters

cols	Column size of A and B.
rowsA	Row size of A.
rankA	Rank of A.
rowsB	Row size of B.
rankA	Rank of B.
rankAB	Rank of [AB].

◆ randn()

Eigen::MatrixXd jrl::qp::test::randn	(	Eigen::Index	rows,
		Eigen::Index	cols,
		Eigen::Index	rank = `-1`
	)

inline

Generate a random matrix with a specified rank.

Coefficients of the matrix have empirically a normal distribution with a mean of 0 and a standard deviation of 1.

Parameters

rows	Number of rows
cols	Number of cols
rank	Rank of the matrix. Must be lower or equal to both `rows` and `cols`. If let to is default value or set negative, the minimum of `rows` and `cols` is taken instead.

◆ randnMat()

auto jrl::qp::test::randnMat	(	Eigen::Index	rows,
		Eigen::Index	cols,
		double	mean = `0`,
		double	stddev = `1`
	)

inline

Generate a random matrix with elements following a normal distribution.

Parameters

rows	Number of rows of the matrix.
cols	Number of columns of the matrix.
mean	Mean parameter of the normal distribution.
stddev	Standard deviation parameter of the normal distribution.

◆ randnVec()

auto jrl::qp::test::randnVec	(	Eigen::Index	size,
		double	mean = `0`,
		double	stddev = `1`
	)

inline

Generate a random vector with elements following a normal distribution.

Parameters

size	Size of the vector.
mean	Mean parameter of the normal distribution.
stddev	Standard deviation parameter of the normal distribution.

◆ randomProblem()

RandomLeastSquare jrl::qp::test::randomProblem ( const ProblemCharacteristics & characs )

◆ randOrtho()

Eigen::MatrixXd jrl::qp::test::randOrtho	(	Eigen::Index	size,
		bool	special = `false`
	)

inline

Generate a random orthogonal matrix with Haar distribution.

Parameters

size	Number of rows and columns of the matrix.
special	If true, the determinant of the generated matrix is 1 (i.e the matrix belongs to SO(size)). Otherwise it is 1 or -1 with 50% probability each (it belongs to O(size))

◆ randUnitVec()

Eigen::VectorXd jrl::qp::test::randUnitVec ( Eigen::Index size )

inline

Generate a random unit vector with uniform distribution on the unit sphere.

Parameters

size	Size of the vector.

◆ testKKT() [1/2]

bool jrl::qp::test::testKKT	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const MatrixConstRef &	G,
		const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`,
		double	tau_p = `1e-6`,
		double	tau_d = `1e-6`
	)

Check the KKT conditions at the dual pair (x,u) for the problem min. .5 x^T G x + a^T x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. u is the vector of Lagrange multipliers with multipliers corresponding to the constraints first and multipliers corresponding to the bound last.

◆ testKKT() [2/2]

bool JRLQP_DLLAPI jrl::qp::test::testKKT	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const QPProblem<> &	pb,
		double	tau_p = `1e-6`,
		double	tau_d = `1e-6`
	)

Overload for a problem given by a QPProblem object.

◆ testKKTFeasibility() [1/2]

bool JRLQP_DLLAPI jrl::qp::test::testKKTFeasibility	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const FeasibilityConstraints &	cstr,
		double	tau_p = `1e-6`,
		double	tau_d = `1e-6`
	)

Overload for a problem given by a QPProblem object.

◆ testKKTFeasibility() [2/2]

bool jrl::qp::test::testKKTFeasibility	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`,
		double	tau_p = `1e-6`,
		double	tau_d = `1e-6`
	)

Check all KKT conditions other than the stationarity at the dual pair (x,u) for the problem min. .5 x^T G x + a^T x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. u is the vector of Lagrange multipliers with multipliers corresponding to the constraints first and multipliers corresponding to the bound last.

◆ testKKTStationarity() [1/2]

bool jrl::qp::test::testKKTStationarity	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const MatrixConstRef &	G,
		const VectorConstRef &	a,
		const MatrixConstRef &	C,
		const VectorConstRef &	bl,
		const VectorConstRef &	bu,
		const VectorConstRef &	xl,
		const VectorConstRef &	xu,
		bool	transposedC = `false`,
		double	tau_d = `1e-6`
	)

Check the stationnarity part of the KKT conditions (i.e. dL/dx = 0, where L is the Lagragian), at the dual pair (x,u) for the problem min. .5 x^T G x + a^T x s.t. bl <= C x <= bu xl <= x <= xu where the variable bounds are optional. u is the vector of Lagrange multipliers with multipliers corresponding to the constraints first and multipliers corresponding to the bound last.

◆ testKKTStationarity() [2/2]

bool JRLQP_DLLAPI jrl::qp::test::testKKTStationarity	(	const VectorConstRef &	x,
		const VectorConstRef &	u,
		const QPProblem<> &	pb,
		double	tau_d = `1e-6`
	)

Overload for a problem given by a QPProblem object.

Variable Documentation

◆ SeparatedFeasibilityConstraints

struct JRLQP_DLLAPI jrl::qp::test::SeparatedFeasibilityConstraints

Classes

Functions

Variables

Function Documentation

◆ checkDimensions() [1/7]

◆ checkDimensions() [2/7]

◆ checkDimensions() [3/7]

◆ checkDimensions() [4/7]

◆ checkDimensions() [5/7]

◆ checkDimensions() [6/7]

◆ checkDimensions() [7/7]

◆ disp()

◆ generateBoxAndSingleConstraintProblem()

◆ randDependent()

◆ randn()

◆ randnMat()

◆ randnVec()

◆ randomProblem()

◆ randOrtho()

◆ randUnitVec()

◆ testKKT() [1/2]

◆ testKKT() [2/2]

◆ testKKTFeasibility() [1/2]

◆ testKKTFeasibility() [2/2]

◆ testKKTStationarity() [1/2]

◆ testKKTStationarity() [2/2]

Variable Documentation

◆ SeparatedFeasibilityConstraints