Collaboration diagram for FullPhysics::NLLSSolverLM::Options:

Public Attributes
double	cr_ratio_tol
	A positive number and tolerance on the ratio of the actual to the predicted reduction of the value of the cost function after taking a step. More...

double	min_W
	A positive number and the smallest value for the diagonal entries of the diagonal weight matrix. More...

double	tr_rad
	A positive number and the initial trust region radius. More...

double	tr_rad_tol
	A positive number and a tolerance for comparing to the trust region radius. More...

Detailed Description

Definition at line 33 of file nlls_solver_lm.h.

Constructor & Destructor Documentation

FullPhysics::NLLSSolverLM::Options::Options ( )

inline

Definition at line 36 of file nlls_solver_lm.h.

double FullPhysics::NLLSSolverLM::Options::cr_ratio_tol

A positive number and tolerance on the ratio of the actual to the predicted reduction of the value of the cost function after taking a step.

A ratio greater than this tolerance results in accepting the step. Otherwise, the step is rejected.

Definition at line 49 of file nlls_solver_lm.h.

double FullPhysics::NLLSSolverLM::Options::min_W

A positive number and the smallest value for the diagonal entries of the diagonal weight matrix.

Definition at line 40 of file nlls_solver_lm.h.

double FullPhysics::NLLSSolverLM::Options::tr_rad

A positive number and the initial trust region radius.

Definition at line 47 of file nlls_solver_lm.h.

double FullPhysics::NLLSSolverLM::Options::tr_rad_tol

A positive number and a tolerance for comparing to the trust region radius.

A step within this tolerance distance from the radius is considered to be on the radius.

Definition at line 43 of file nlls_solver_lm.h.

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