eoOneFifthMutation< EOT > Class Template Reference

the dynamic version: just say it is updatable - and write the update() method! here the 1 fifth rule: count the proportion of successful mutations, and increase sigma if more than threshold (1/5 !) More...

#include <eoNormalMutation.h>

Inheritance diagram for eoOneFifthMutation< EOT >:

List of all members.


Public Types
typedef EOT::Fitness	Fitness
Public Member Functions
	eoOneFifthMutation (eoEvalFunc< EOT > &_eval, double &_sigmaInit, unsigned _windowSize=10, double _updateFactor=0.83, double _threshold=0.2)
	(Default) Constructor.
virtual std::string	className () const
	The class name.
bool	operator() (EOT &_eo)
	Do it! calls the standard mutation, then checks for success and updates stats.
void	update ()
	the method that will be called every generation if the object is added to the checkpoint
Private Attributes
eoEvalFunc< EOT > &	eval
double	threshold
double	updateFactor
std::vector< unsigned >	nbMut
std::vector< unsigned >	nbSuccess
unsigned	genIndex

Detailed Description

template<class EOT>
class eoOneFifthMutation< EOT >

the dynamic version: just say it is updatable - and write the update() method! here the 1 fifth rule: count the proportion of successful mutations, and increase sigma if more than threshold (1/5 !)

Definition at line 177 of file eoNormalMutation.h.

Constructor & Destructor Documentation

template<class EOT>

eoOneFifthMutation< EOT >::eoOneFifthMutation	(	eoEvalFunc< EOT > &	_eval,
		double &	_sigmaInit,
		unsigned	_windowSize = `10`,
		double	_updateFactor = `0.83`,
		double	_threshold = `0.2`
	)			`[inline]`

(Default) Constructor.

Parameters:

	eval	the evaluation function, needed to recompute the fitmess
	_sigmaInit	the initial value for uniform mutation
	_windowSize	the size of the window for statistics
	_threshold	the threshold (the 1/5 - 0.2)
	_updateFactor	multiplicative update factor for sigma