Optimisation

Classes
class	MR::Math::QuadraticLineSearch< ValueType >
	Computes the minimum of a 1D function using a quadratic line search. More...

Functions
template<class FunctionType , typename ValueType >
ValueType	MR::Math::golden_section_search (FunctionType &function, const std::string &message, ValueType min_bound, ValueType init_estimate, ValueType max_bound, ValueType tolerance=0.01)
	Computes the minimum of a 1D function using a golden section search. More...

Detailed Description

Function Documentation

golden_section_search()

template<class FunctionType , typename ValueType >

ValueType MR::Math::golden_section_search	(	FunctionType &	function,
		const std::string &	message,
		ValueType	min_bound,
		ValueType	init_estimate,
		ValueType	max_bound,
		ValueType	tolerance = 0.01
	)

Computes the minimum of a 1D function using a golden section search.

This function operates on a cost function class that must define a operator() method. The method must take a single ValueType argument x and return the cost of the function at x.

The min_bound and max_bound arguments define values that bracket the expected minimum. The estimate argument is the initial estimate of the minimum that is required to be larger than min_bound and smaller than max_bound.

Typical usage:

CostFunction cost_function();
float optimal_value = Math::golden_section_search(cost_function, "optimising", min_bound, initial_estimate , max_bound);

Definition at line 50 of file golden_section_search.h.