Applies conic operations to each cone in the solver. More...

Functions
int	DSDPAddCone (DSDP dsdp, struct DSDPCone_Ops dsdpops, void dsdpcone)
	Apply DSDP to a conic structure. More...

int	DSDPAddSchurRow (DSDP, int, DSDPVec)
	Add a row to the Schur matrix. More...

int	DSDPComputeANorm2 (DSDP dsdp, DSDPVec Anorm2)
	Compute norm of A and C. More...

int	DSDPComputeG (DSDP dsdp, DSDPVec vt, DSDPVec vrhs1, DSDPVec vrhs2)
	Compute the gradient of the barrier for each cone. More...

int	DSDPComputeHessian (DSDP dsdp, DSDPSchurMat M, DSDPVec vrhs1, DSDPVec vrhs2)
	Compute the Schur complement, or Gram, matrix for each cone. More...

int	DSDPComputeLogSDeterminant (DSDP dsdp, double *logdet)
	Compute the logarithmic barrier function for the dual varialbe S. More...

int	DSDPComputeMaxStepLength (DSDP dsdp, DSDPVec DY, DSDPDualFactorMatrix flag, double *maxsteplength)
	Compute the maximum step length for the given step direction. More...

int	DSDPComputeSS (DSDP dsdp, DSDPVec Y, DSDPDualFactorMatrix flag, DSDPTruth *ispsdefinite)
	Compute the dual variables S in each cone. More...

int	DSDPComputeXVariables (DSDP dsdp, double xmakermu, DSDPVec xmakery, DSDPVec xmakerdy, DSDPVec AX, double *tracexs)
	Compute the X variables in each cone. More...

int	DSDPDestroyCones (DSDP dsdp)
	Each cone shoudl free its data structures. More...

int	DSDPGetConicDimension (DSDP dsdp, double *n)
	Get the total dimension of the cones. More...

int	DSDPHessianMultiplyAdd (DSDP dsdp, DSDPVec v, DSDPVec vv)
	Add the product of Schur matrix with v. More...

int	DSDPInvertS (DSDP dsdp)
	Invert the S variables in each cone. More...

int	DSDPMonitorCones (DSDP dsdp, int tag)
	This routine is called once per iteration. More...

int	DSDPPassXVectors (DSDP dsdp, double mu, DSDPVec Y, DSDPVec DY)
	Pass the information needed to compute the variables X in each cone but do not compute X. More...

int	DSDPSchurSparsity (DSDP dsdp, int row, int rnnz[], int m)
	Each cone should print its state. More...

int	DSDPSetCone (DSDP dsdp, DSDPCone tcone)
	Pass a cone to the DSDP solver. More...

int	DSDPSetSchurMatOps (DSDP, struct DSDPSchurMat_Ops , void )
	Set the Schur complement matrix. More...

int	DSDPSetUpCones (DSDP dsdp)
	Each cone should factor data or allocate internal data structures. More...

int	DSDPSetUpCones2 (DSDP dsdp, DSDPVec yy0, DSDPSchurMat M)
	Each cone should allocate its data structures . More...

int	DSDPViewCones (DSDP dsdp)
	Each cone should print its state. More...

Detailed Description

Applies conic operations to each cone in the solver.

Definition in file dsdpcops.c.

Function Documentation

◆ DSDPAddCone()

int DSDPAddCone	(	DSDP	dsdp,
		struct DSDPCone_Ops *	dsdpops,
		void *	dsdpcone
	)

Apply DSDP to a conic structure.

Parameters

dsdp	the solver
dsdpops	address of a structure with function pointers
dsdpcone	address of a cone structure DSDP operates on cones such as the semidefinite cone and nonnegative orthant. Given variables y from the solver, each cone implements operations such as computing S, maximum step length, computing the Newton matrix, and computing the Hessian. Each operation is well defined by the dual-scaling algorithm. A cone that implements these operations can be added to the DSDP solver.

See also: DSDPCreateSDPCone(); DSDPCreateLPCone(); DSDPCreateBCone()

Todo:: Add SOCP cone and application-specific cones.

Definition at line 569 of file dsdpcops.c.

◆ DSDPAddSchurRow()

int DSDPAddSchurRow	(	DSDP	dsdp,
		int	row,
		DSDPVec	R
	)

Add a row to the Schur matrix.

Parameters

dsdp	the solver
row	corresponding to which variable y.
R	the elements of the row.

This routine is called by the conic routines that compute the Hessian matrix.

Definition at line 622 of file dsdpcops.c.

◆ DSDPComputeANorm2()

int DSDPComputeANorm2	(	DSDP	dsdp,
		DSDPVec	Anorm2
	)

Compute norm of A and C.

Parameters

dsdp	the solver
Anorm2	norm of data corresponding to each variable y.

Definition at line 246 of file dsdpcops.c.

Referenced by DSDPComputeDataNorms().

◆ DSDPComputeG()

int DSDPComputeG	(	DSDP	dsdp,
		DSDPVec	vt,
		DSDPVec	vrhs1,
		DSDPVec	vrhs2
	)

Compute the gradient of the barrier for each cone.

Parameters

dsdp	the solver
vt	scaling for each element in the next two vectors
vrhs1	scaled gradient of the objective function
vrhs2	scaled gradient of the barrier function

Definition at line 215 of file dsdpcops.c.

Referenced by DSDPComputeDualStepDirections().

◆ DSDPComputeHessian()

int DSDPComputeHessian	(	DSDP	dsdp,
		DSDPSchurMat	M,
		DSDPVec	vrhs1,
		DSDPVec	vrhs2
	)

Compute the Schur complement, or Gram, matrix for each cone.

Parameters

dsdp	the solver
M	matrix
vrhs1	gradient of objective (b)
vrhs2	gradient of barrier

Definition at line 142 of file dsdpcops.c.

Referenced by DSDPComputeDualStepDirections().

◆ DSDPComputeLogSDeterminant()

int DSDPComputeLogSDeterminant	(	DSDP	dsdp,
		double *	logdet
	)

Compute the logarithmic barrier function for the dual varialbe S.

Parameters

dsdp	the solver
logdet	evaluated barrier function

See also: DSDPComputeSS()

Definition at line 495 of file dsdpcops.c.

Referenced by DSDPYStepLineSearch().

◆ DSDPComputeMaxStepLength()

int DSDPComputeMaxStepLength	(	DSDP	dsdp,
		DSDPVec	DY,
		DSDPDualFactorMatrix	flag,
		double *	maxsteplength
	)

Compute the maximum step length for the given step direction.

Parameters

dsdp	the solver
DY	step direction
flag	primal or dual structure
maxsteplength	the minumum of maximums on each cone.

See also: DSDPComputeSS()

Definition at line 336 of file dsdpcops.c.

Referenced by DSDPChooseBarrierParameter(), DSDPYStepLineSearch(), and DSDPYStepLineSearch2().

◆ DSDPComputeSS()

int DSDPComputeSS	(	DSDP	dsdp,
		DSDPVec	Y,
		DSDPDualFactorMatrix	flag,
		DSDPTruth *	ispsdefinite
	)

Compute the dual variables S in each cone.

Parameters

dsdp	the solver
Y	variables
flag	primal or dual structure
ispsdefinite	DSDP_TRUE if a member of the cone, DSDP_FALSE otherwise.

Definition at line 272 of file dsdpcops.c.

Referenced by DSDPChooseBarrierParameter(), DSDPComputeAndFactorS(), DSDPInitializeVariables(), and DSDPYStepLineSearch().

◆ DSDPComputeXVariables()

int DSDPComputeXVariables	(	DSDP	dsdp,
		double	xmakermu,
		DSDPVec	xmakery,
		DSDPVec	xmakerdy,
		DSDPVec	AX,
		double *	tracexs
	)

Compute the X variables in each cone.

Parameters

dsdp	the solver
xmakermu	barrier parameter
xmakery	input y variables
xmakerdy	input step direction
AX	output product of X and the data
tracexs	ouput inner product of X and S.

Definition at line 654 of file dsdpcops.c.

◆ DSDPDestroyCones()

int DSDPDestroyCones ( DSDP dsdp )

Each cone shoudl free its data structures.

Parameters

dsdp	the solver

Definition at line 107 of file dsdpcops.c.

◆ DSDPGetConicDimension()

int DSDPGetConicDimension	(	DSDP	dsdp,
		double *	n
	)

Get the total dimension of the cones.

Parameters

dsdp	the solver
n	dimension

Definition at line 401 of file dsdpcops.c.

Referenced by DSDPCheckConvergence(), and DSDPGetDimension().

◆ DSDPHessianMultiplyAdd()

int DSDPHessianMultiplyAdd	(	DSDP	dsdp,
		DSDPVec	v,
		DSDPVec	vv
	)

Add the product of Schur matrix with v.

Parameters

dsdp	the solver
v	input vector.
vv	product gradient of barrier

Definition at line 188 of file dsdpcops.c.

◆ DSDPInvertS()

int DSDPInvertS ( DSDP dsdp )

Invert the S variables in each cone.

Parameters

dsdp	the solver

See also: DSDPComputeSS()

Definition at line 307 of file dsdpcops.c.

Referenced by DSDPComputeDualStepDirections().

◆ DSDPMonitorCones()

int DSDPMonitorCones	(	DSDP	dsdp,
		int	tag
	)

This routine is called once per iteration.

Parameters

dsdp	the solver
tag	allow for multiple monitors

The cone can print statistics, visualize data, terminate solver, or whatever it wants.

Definition at line 450 of file dsdpcops.c.

◆ DSDPPassXVectors()

int DSDPPassXVectors	(	DSDP	dsdp,
		double	mu,
		DSDPVec	Y,
		DSDPVec	DY
	)

Pass the information needed to compute the variables X in each cone but do not compute X.

Parameters

dsdp	the solver
mu	barrier parameter
Y	input y variables
DY	input step direction

See also: DSDPComputeXVariables()

Definition at line 378 of file dsdpcops.c.

◆ DSDPSchurSparsity()

int DSDPSchurSparsity	(	DSDP	dsdp,
		int	row,
		int	rnnz[],
		int	m
	)

Each cone should print its state.

Parameters

dsdp	the solver
row	corresponding to the variable y.
rnnz	nonzeros indicate a nonzero in the Shur matrix at that column.
m	size of Schur matrix and the arrow.

Definition at line 474 of file dsdpcops.c.

◆ DSDPSetCone()

int DSDPSetCone	(	DSDP	dsdp,
		DSDPCone	tcone
	)

Pass a cone to the DSDP solver.

Parameters

dsdp	the solver
tcone	a cone object.

Definition at line 522 of file dsdpcops.c.

Referenced by DSDPAddCone().

◆ DSDPSetSchurMatOps()

int DSDPSetSchurMatOps	(	DSDP	dsdp,
		struct DSDPSchurMat_Ops *	sops,
		void *	mdata
	)

Set the Schur complement matrix.

Parameters

dsdp	the solver
sops	address of a structure with function pointers
mdata	address of a matrix object The step direction in DSDP is the solution to a set of linear equations. The cones used by DSDP compute the elements of the matrix and the right-hand side vectors. Any matrix that implements the Schur complement matrix interface can be used by DSDP. In addition to factoring a matrix and solving it, this interface also provides matrix assembly routines for the cones.

See also: DSDPAddCone()

Todo:: Use SCALAPACK to assemble, factor, and solve the matrix in parallel.

Definition at line 602 of file dsdpcops.c.

◆ DSDPSetUpCones()

int DSDPSetUpCones ( DSDP dsdp )

Each cone should factor data or allocate internal data structures.

Parameters

dsdp	the solver

Definition at line 58 of file dsdpcops.c.

◆ DSDPSetUpCones2()

int DSDPSetUpCones2	(	DSDP	dsdp,
		DSDPVec	yy0,
		DSDPSchurMat	M
	)

Each cone should allocate its data structures .

Parameters

dsdp	the solver
yy0	variable vector
M	Shur Matrix

Definition at line 84 of file dsdpcops.c.

◆ DSDPViewCones()

int DSDPViewCones ( DSDP dsdp )

Each cone should print its state.

Parameters

dsdp	the solver

Definition at line 424 of file dsdpcops.c.

Functions

Detailed Description

Function Documentation

◆ DSDPAddCone()

◆ DSDPAddSchurRow()

◆ DSDPComputeANorm2()

◆ DSDPComputeG()

◆ DSDPComputeHessian()

◆ DSDPComputeLogSDeterminant()

◆ DSDPComputeMaxStepLength()

◆ DSDPComputeSS()

◆ DSDPComputeXVariables()

◆ DSDPDestroyCones()

◆ DSDPGetConicDimension()

◆ DSDPHessianMultiplyAdd()

◆ DSDPInvertS()

◆ DSDPMonitorCones()

◆ DSDPPassXVectors()

◆ DSDPSchurSparsity()

◆ DSDPSetCone()

◆ DSDPSetSchurMatOps()

◆ DSDPSetUpCones()

◆ DSDPSetUpCones2()

◆ DSDPViewCones()