nojhan/paradiseo
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Add the DUAL fitness assignment class

Browse source 
Change a bit the machinery of moeoExpBinaryIndicatorBasedFitnessAssignment to allow subclassing.
This commit is contained in:
Johann Dreo 2013-06-07 17:50:04 +02:00
commit 45123abbf3
3 changed files with 76 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

67
moeo/src/fitness/moeoExpBinaryIndicatorBasedDualFitnessAssignment.h Normal file
View file

@ -0,0 +1,67 @@
#include <fitness/moeoExpBinaryIndicatorBasedFitnessAssignment.h>
template<class MOEOT>
class moeoExpBinaryIndicatorBasedDualFitnessAssignment : public moeoExpBinaryIndicatorBasedFitnessAssignment<MOEOT>
{
protected:
eoPop<MOEOT> _feasible_pop;
eoPop<MOEOT> _unfeasible_pop;
public:
typedef typename MOEOT::ObjectiveVector ObjectiveVector;
moeoExpBinaryIndicatorBasedDualFitnessAssignment(
moeoNormalizedSolutionVsSolutionBinaryMetric<ObjectiveVector,double> & metric,
const double kappa = 0.05
) : moeoExpBinaryIndicatorBasedFitnessAssignment<MOEOT>( metric, kappa ) {}
//! Split up the population in two: in one pop the feasible individual, in the other the feasible ones
virtual void split( eoPop<MOEOT> & pop )
{
_feasible_pop.reserve(pop.size());
_unfeasible_pop.reserve(pop.size());
for( typename eoPop<MOEOT>::iterator it=pop.begin(), end=pop.end(); it != end; ++it ) {
// The ObjectiveVector should implement "is_feasible"
if( it->objectiveVector().is_feasible() ) {
_feasible_pop.push_back( *it );
} else {
_unfeasible_pop.push_back( *it );
}
}
}
/*! If the population is homogeneous (only composed of feasible individuals or unfeasible ones),
* then apply the operators on the whole population.
* But, if there is at least one feasible individual, then apply them only on the feasible individuals.
*/
virtual void operator()(eoPop < MOEOT > & pop)
{
// separate the pop in the members
split( pop );
eoPop<MOEOT>* ppop;
// if there is at least one feasible individual, it will supersede all the unfeasible ones
if( _feasible_pop.size() == 0 ) {
ppop = & _unfeasible_pop;
} else {
ppop = & _feasible_pop;
}
this->setup(*ppop);
this->computeValues(*ppop);
this->setFitnesses(*ppop);
}
virtual void setFitnesses(eoPop < MOEOT > & pop)
{
for (unsigned int i=0; i<pop.size(); i++) {
// We should maintain the feasibility of the fitness across computations
pop[i].fitness( this->computeFitness(i), pop[i].is_feasible() );
}
}
};