qpalm::Settings Struct Reference

#include <qpalm.hpp>

Detailed Description

Settings and parameters for the QPALM solver.

Examples

examples/cxx/qpalm_demo.cpp.

Definition at line 110 of file qpalm.hpp.

Inheritance diagram for qpalm::Settings:

Collaboration diagram for qpalm::Settings:

Public Member Functions
	Settings ()
	Construct with default settings.

Data Fields
c_int	max_iter
	maximum number of iterations
c_int	inner_max_iter
	maximum number of iterations per subproblem
c_float	eps_abs
	absolute convergence tolerance
c_float	eps_rel
	relative convergence tolerance
c_float	eps_abs_in
	intermediate absolute convergence tolerance
c_float	eps_rel_in
	intermediate relative convergence tolerance
c_float	rho
	tolerance scaling factor
c_float	eps_prim_inf
	primal infeasibility tolerance
c_float	eps_dual_inf
	dual infeasibility tolerance
c_float	theta
	penalty update criterion parameter
c_float	delta
	penalty update factor
c_float	sigma_max
	penalty factor cap
c_float	sigma_init
	initial penalty parameter (guideline)
c_int	proximal
	boolean, use proximal method of multipliers or not
c_float	gamma_init
	initial proximal penalty parameter
c_float	gamma_upd
	proximal penalty update factor
c_float	gamma_max
	proximal penalty parameter cap
c_int	scaling
	scaling iterations, if 0 then scaling is disabled
c_int	nonconvex
	boolean, indicates whether the QP is nonconvex
c_int	verbose
	boolean, write out progress
c_int	print_iter
	frequency of printing
c_int	warm_start
	boolean, warm start
c_int	reset_newton_iter
	frequency of performing a complete Cholesky factorization
c_int	enable_dual_termination
	boolean, enable termination based on dual objective (useful in branch and bound)
c_float	dual_objective_limit
	termination value for the dual objective (useful in branch and bound)
c_float	time_limit
	time limit
c_int	ordering
	ordering method for factorization
c_int	factorization_method
	factorize KKT or Schur complement
c_int	max_rank_update
	maximum rank for the sparse factorization update
c_float	max_rank_update_fraction
	maximum rank (relative to n+m) for the factorization update

Constructor & Destructor Documentation

Settings()

qpalm::Settings::Settings

(

)

Construct with default settings.

See also

qpalm_set_default_settings

Definition at line 21 of file qpalm.cpp.

Here is the call graph for this function:

Field Documentation

max_iter

c_int QPALMSettings::max_iter

inherited

maximum number of iterations

Note

Assumption: \(>0\)

Examples

examples/cxx/qpalm_demo.cpp.

Definition at line 118 of file types.h.

inner_max_iter

c_int QPALMSettings::inner_max_iter

inherited

maximum number of iterations per subproblem

Note

Assumption: \(>0\)

Definition at line 119 of file types.h.

eps_abs

c_float QPALMSettings::eps_abs

inherited

absolute convergence tolerance

Note

Assumption: \(>=0\), either eps_abs or eps_rel must be \(>0\)

Examples

examples/cxx/qpalm_demo.cpp.

Definition at line 120 of file types.h.

eps_rel

c_float QPALMSettings::eps_rel

inherited

relative convergence tolerance

Note

Assumption: \(>=0\), either eps_abs or eps_rel must be \(>0\)

Examples

examples/cxx/qpalm_demo.cpp.

Definition at line 121 of file types.h.

eps_abs_in

c_float QPALMSettings::eps_abs_in

inherited

intermediate absolute convergence tolerance

Note

Assumption: \(>=0\), either eps_abs_in or eps_rel_in must be \(>0\)

Definition at line 122 of file types.h.

eps_rel_in

c_float QPALMSettings::eps_rel_in

inherited

intermediate relative convergence tolerance

Note

Assumption: \(>=0\), either eps_abs_in or eps_rel_in must be \(>0\)

Definition at line 123 of file types.h.

rho

c_float QPALMSettings::rho

inherited

tolerance scaling factor

Note

Assumption: \(0<\rho<1\)

Definition at line 124 of file types.h.

eps_prim_inf

c_float QPALMSettings::eps_prim_inf

inherited

primal infeasibility tolerance

Note

Assumption: \(>=0\)

Definition at line 125 of file types.h.

eps_dual_inf

c_float QPALMSettings::eps_dual_inf

inherited

dual infeasibility tolerance

Note

Assumption: \(>=0\)

Definition at line 126 of file types.h.

theta

c_float QPALMSettings::theta

inherited

penalty update criterion parameter

Note

Assumption: \(<=1\)

Definition at line 127 of file types.h.

delta

c_float QPALMSettings::delta

inherited

penalty update factor

Note

Assumption: \(>1\)

Definition at line 128 of file types.h.

sigma_max

c_float QPALMSettings::sigma_max

inherited

penalty factor cap

Note

Assumption: \(>0\)

Definition at line 129 of file types.h.

sigma_init

c_float QPALMSettings::sigma_init

inherited

initial penalty parameter (guideline)

Note

Assumption: \(>0\)

Definition at line 130 of file types.h.

proximal

c_int QPALMSettings::proximal

inherited

boolean, use proximal method of multipliers or not

Note

Assumption: \(\in \{0,1\}\)

Definition at line 131 of file types.h.

gamma_init

c_float QPALMSettings::gamma_init

inherited

initial proximal penalty parameter

Note

Assumption: \(>0\)

Definition at line 132 of file types.h.

gamma_upd

c_float QPALMSettings::gamma_upd

inherited

proximal penalty update factor

Note

Assumption: \(>=1\)

Definition at line 133 of file types.h.

gamma_max

c_float QPALMSettings::gamma_max

inherited

proximal penalty parameter cap

Note

Assumption: \(>=\gamma_\textrm{init}\)

Definition at line 134 of file types.h.

scaling

c_int QPALMSettings::scaling

inherited

scaling iterations, if 0 then scaling is disabled

Note

Assumption: \(>=0\)

Definition at line 135 of file types.h.

nonconvex

c_int QPALMSettings::nonconvex

inherited

boolean, indicates whether the QP is nonconvex

Note

Assumption: \(\in \{0,1\}\)

Definition at line 136 of file types.h.

verbose

c_int QPALMSettings::verbose

inherited

boolean, write out progress

Note

Assumption: \(\in \{0,1\}\)

Definition at line 137 of file types.h.

print_iter

c_int QPALMSettings::print_iter

inherited

frequency of printing

Note

Assumption: \(>0\)

Definition at line 138 of file types.h.

warm_start

c_int QPALMSettings::warm_start

inherited

boolean, warm start

Note

Assumption: \(\in \{0,1\}\)

Definition at line 139 of file types.h.

reset_newton_iter

c_int QPALMSettings::reset_newton_iter

inherited

frequency of performing a complete Cholesky factorization

Note

Assumption: \(>0\)

Definition at line 140 of file types.h.

enable_dual_termination

c_int QPALMSettings::enable_dual_termination

inherited

boolean, enable termination based on dual objective (useful in branch and bound)

Note

Assumption: \(\in \{0,1\}\)

Definition at line 141 of file types.h.

dual_objective_limit

c_float QPALMSettings::dual_objective_limit

inherited

termination value for the dual objective (useful in branch and bound)

Note

Assumption: none

Definition at line 142 of file types.h.

time_limit

c_float QPALMSettings::time_limit

inherited

time limit

Note

Assumption: \(>0\)

Definition at line 143 of file types.h.

ordering

c_int QPALMSettings::ordering

inherited

ordering method for factorization

Definition at line 144 of file types.h.

factorization_method

c_int QPALMSettings::factorization_method

inherited

factorize KKT or Schur complement

Definition at line 145 of file types.h.

max_rank_update

c_int QPALMSettings::max_rank_update

inherited

maximum rank for the sparse factorization update

Definition at line 146 of file types.h.

max_rank_update_fraction

c_float QPALMSettings::max_rank_update_fraction

inherited

maximum rank (relative to n+m) for the factorization update

Definition at line 147 of file types.h.

---

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

• QPALM/interfaces/cxx/include/qpalm.hpp
• QPALM/interfaces/cxx/src/qpalm.cpp