syn::synthetic Class Reference

#include <synthetic.hpp>

Inheritance diagram for syn::synthetic:

Public Member Functions
	synthetic (const std::string &n, u_int dims, double lb, double ub, double opt)
	synthetic (const std::string &n, u_int dims, double opt)
	~synthetic ()
sspace::sspace_t *	get_search_space ()
void	set_properties (std::vector< properties > p)
std::vector< properties >	get_properties ()
inst::set	get_opt_param ()
double	get_opt ()
void	update_opt (double opt)
u_int	get_dims ()
void	validate_param_set (inst::set x)
Public Member Functions inherited from optk::benchmark
	benchmark (const std::string &name)
std::string	get_name ()
virtual double	evaluate (inst::set x)=0

Protected Member Functions
void	set_opt_param (inst::set op)

Protected Attributes
u_int	m_dims
double	m_lb
double	m_ub
double	m_opt
std::vector< properties >	m_properties
inst::set	opt_params
sspace::sspace_t	m_sspace
Protected Attributes inherited from optk::benchmark
std::string	m_name

Detailed Description

Owing to the similarity of the structure, and subsequent exposition, of the 175 functions in Jamil et al. 2013, we use this as base class for the synthetic benchmarks.

Constructor & Destructor Documentation

synthetic() [1/2]

syn::synthetic::synthetic	(	const std::string &	n,
		u_int	dims,
		double	lb,
		double	ub,
		double	opt
	)

Base constructor for synthetic benchmarks. Unless overidden by derived calsses, this will expand the provided lower bound, upper bound and dimensionality to a uniform bounded parameter space of type sspace::sspace_t, making use of heap memory allocations.

Parameters

n	The name of the benchmark.
dims	The number of dimensions in which to run this benchmark.
lb	The lower bound on the parameters constituting the search space.
up	The corresponding upper bound.
opt	The global minimum for this function.
opt_param	The optimal parameters for this function.

synthetic() [2/2]

syn::synthetic::synthetic	(	const std::string &	n,
		u_int	dims,
		double	opt
	)

A simpler base constructor for synthetic benchmarks which have more complicated search spaces which will be manually set in the derived class' constructor.

Parameters

n	The name of the benchmark
dims	The number of dimensions of this benchmark
opt	The global minimum

~synthetic()

syn::synthetic::~synthetic

(

)

This destructor frees the search space which was automatically-generated in the constructor.

Member Function Documentation

get_dims()

u_int syn::synthetic::get_dims ( )

inline

Get the number of dimensions that this problem was instantiated with; or the fixed dimension that the derived function can take.

get_opt()

double syn::synthetic::get_opt ( )

inline

Returns (one of) the known global minima.

get_opt_param()

inst::set syn::synthetic::get_opt_param ( )

inline

Returns the known optimal parameter set.

get_search_space()

sspace::sspace_t* syn::synthetic::get_search_space ( )

inlinevirtual

Returns

The search space for this benchmark. This is passed to the update_search_space method of optimisation algorithms.

Implements optk::benchmark.

set_opt_param()

void syn::synthetic::set_opt_param ( inst::set  op )

protected

Sets the parameters which give the optimum when evaluated. This set needn't be unique; for instance in multimodal functions it is not.

Only call once from the constructor of a derived class; multiple invocations leads to memory leaks. It is 'protected' to guard against this.

Parameters

op	The set of optimal parameter instances.

set_properties()

void syn::synthetic::set_properties ( std::vector< properties >  p )

inline

This is specific to synthetic benchmarks, and sets the properties of the derived benchmark function (e.g. non-separable, scalable etc...)

Parameters

p	A vector of elements from the properties enum.

update_opt()

void syn::synthetic::update_opt ( double  opt )

inline

Update the value of the global minimum.

validate_param_set()

void syn::synthetic::validate_param_set

(

inst::set

x

)

This function validates that a set of parameter instances provided to be evaluated is compatible with the search space.

Parameters

x	The search space provided by the optimisation algorithm.

Exceptions

std::invalid_argument

when the parameter set instance is invalid.

---

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

• includes/benchmarks/synthetic.hpp
• src/benchmarks/synthetic.cpp