Wrapper class for GSL gradient - based multidimensional function minimization routines. More...
#include <src/gsl_wrappers/gsl_multidim_fdfminimizer.h>
Public Member Functions | |
multimin_fdfminimizer (uint _dim, const gsl_multimin_fdfminimizer_type *type=gsl_multimin_fdfminimizer_vector_bfgs) | |
Class constructor to initialize minimizer. | |
~multimin_fdfminimizer () | |
Class destructor for multimin_fdfminimizer. | |
int | set (multimin_fdf &function, const vector &initial_x, double step_size, double tol) |
Function to set, or reset an existing minimizer. | |
int | iterate () |
Function that performs an iteration of the minimizer. | |
int | restart () |
Function to reset the minimizer to use the current point as a new starting point. | |
double | minimum () |
Function returning current minimum value of function after an iteration. | |
vector | x_value () |
Function that returns the current vector where minimum of the function is attained. | |
vector | gradient () |
Function that returns the gradient vector at the current minimum of the function. | |
void | SetMaxIterations (int n) |
Function that sets the maximum number of iterations. | |
int | GetNIterations () |
Function that returns the number of iteration that were performed. | |
bool | is_converged () |
Function that checks if iterations have converged. | |
Private Member Functions | |
void | assert_set () |
Error handling function. | |
Private Attributes | |
uint | dim |
Unsigned integer for dimension of space where function is defined. | |
bool | isSet |
Boolean variable to check everything is set properly. | |
bool | isConverged |
Boolean variable to check if iterations converge. | |
int | nIterations |
Index for current iteration number. | |
int | maxIterations |
Maximum number of iterations to be pursued. | |
gsl_multimin_fdfminimizer * | s |
GSL multi-dimensional function-derivative minimizer object. | |
gsl_multimin_function_fdf | f |
Multidimensionla function and derivative called by GSL during minimization. |
Wrapper class for GSL gradient - based multidimensional function minimization routines.