ROOT logo
ROOT » TMVA » TMVA::GeneticPopulation

class TMVA::GeneticPopulation


 Population definition for genetic algorithm

Function Members (Methods)

public:
virtual~GeneticPopulation()
voidAddPopulation(TMVA::GeneticPopulation* strangers)
voidAddPopulation(TMVA::GeneticPopulation& strangers)
static TClass*Class()
TMVA::GeneticPopulationGeneticPopulation(const TMVA::GeneticPopulation&)
TMVA::GeneticPopulationGeneticPopulation(const vector<TMVA::Interval*>& ranges, Int_t size, UInt_t seed = 0)
Double_tGetFitness() const
const vector<TMVA::GeneticGenes>&GetGenePool() const
vector<TMVA::GeneticGenes>&GetGenePool()
TMVA::GeneticGenes*GetGenes(Int_t index)
Int_tGetPopulationSize() const
const vector<TMVA::GeneticRange*>&GetRanges() const
vector<TMVA::GeneticRange*>&GetRanges()
voidGiveHint(vector<Double_t>& hint, Double_t fitness = 0)
virtual TClass*IsA() const
voidMakeChildren()
voidMakeCopies(int number)
voidMutate(Double_t probability = 20, Int_t startIndex = 0, Bool_t near = kFALSE, Double_t spread = 0.1, Bool_t mirror = kFALSE)
voidNextGeneration()
TMVA::GeneticPopulation&operator=(const TMVA::GeneticPopulation&)
voidPrint(Int_t untilIndex = -1)
voidPrint(ostream& out, Int_t utilIndex = -1)
voidSetRandomSeed(UInt_t seed = 0)
virtual voidShowMembers(TMemberInspector& insp, char* parent)
voidSort()
virtual voidStreamer(TBuffer& b)
voidStreamerNVirtual(TBuffer& b)
voidTrimPopulation()
vector<Double_t>VariableDistribution(Int_t varNumber)
TH1F*VariableDistribution(Int_t varNumber, Int_t bins, Int_t min, Int_t max)

Data Members

private:
vector<TMVA::GeneticGenes>fGenePoolthe "genePool" where the individuals of the current generation are stored
TMVA::MsgLogger*fLoggermessage logger
Int_tfPopulationSizeLimit
TRandom3*fRandomGeneratorrandom Generator for this population
vector<TMVA::GeneticRange*>fRangescontains the ranges inbetween the values of the coefficients have to be

Class Charts

Inheritance Inherited Members Includes Libraries
Class Charts

Function documentation

GeneticPopulation(const vector<TMVA::Interval*>& ranges, Int_t size, UInt_t seed = 0)
 Constructor
~GeneticPopulation()
 destructor
void SetRandomSeed(UInt_t seed = 0)
 the random seed of the random generator
void MakeCopies(int number)
 produces offspring which is are copies of their parents
 Parameters:
         int number : the number of the last individual to be copied

void MakeChildren()
 does what the name says,... it creates children out of members of the
 current generation
 children have a combination of the coefficients of their parents

TMVA::GeneticGenes MakeSex(TMVA::GeneticGenes male, TMVA::GeneticGenes female)
 this function takes two individuals and produces offspring by mixing (recombining) their
 coefficients

void Mutate(Double_t probability = 20, Int_t startIndex = 0, Bool_t near = kFALSE, Double_t spread = 0.1, Bool_t mirror = kFALSE)
 mutates the individuals in the genePool
 Parameters:
         double probability : gives the probability (in percent) of a mutation of a coefficient
         int startIndex : leaves unchanged (without mutation) the individuals which are better ranked
                     than indicated by "startIndex". This means: if "startIndex==3", the first (and best)
                     three individuals are not mutaded. This allows to preserve the best result of the
                     current Generation for the next generation.
         Bool_t near : if true, the mutation will produce a new coefficient which is "near" the old one
                     (gaussian around the current value)
         double spread : if near==true, spread gives the sigma of the gaussian
         Bool_t mirror : if the new value obtained would be outside of the given constraints
                    the value is mapped between the constraints again. This can be done either
                    by a kind of periodic boundary conditions or mirrored at the boundary.
                    (mirror = true seems more "natural")

TMVA::GeneticGenes* GetGenes(Int_t index)
 gives back the "Genes" of the population with the given index.

void Print(Int_t untilIndex = -1)
 make a little printout of the individuals up to index "untilIndex"
 this means, .. write out the best "untilIndex" individuals.

void Print(ostream& out, Int_t utilIndex = -1)
 make a little printout to the stream "out" of the individuals up to index "untilIndex"
 this means, .. write out the best "untilIndex" individuals.

TH1F* VariableDistribution(Int_t varNumber, Int_t bins, Int_t min, Int_t max)
 give back a histogram with the distribution of the coefficients
 parameters:
          int bins : number of bins of the histogram
          int min : histogram minimum
          int max : maximum value of the histogram

vector<Double_t> VariableDistribution(Int_t varNumber)
 gives back all the values of coefficient "varNumber" of the current generation

void AddPopulation( GeneticPopulation *strangers )
 add another population (strangers) to the one of this GeneticPopulation
void AddPopulation( GeneticPopulation &strangers )
 add another population (strangers) to the one of this GeneticPopulation
void TrimPopulation()
 trim the population to the predefined size
void GiveHint(vector<Double_t>& hint, Double_t fitness = 0)
 add an individual (a set of variables) to the population
 if there is a set of variables which is known to perform good, they can be given as a hint to the population
void Sort()
 sort the genepool according to the fitness of the individuals
GeneticPopulation(const vector<TMVA::Interval*>& ranges, Int_t size, UInt_t seed = 0)
Int_t GetPopulationSize() const
{ return fGenePool.size(); }
Double_t GetFitness() const
{ return fGenePool.size()>0? fGenePool[0].GetFitness() : 0; }
const std::vector<TMVA::GeneticGenes>& GetGenePool()
{ return fGenePool; }
const std::vector<TMVA::GeneticRange*>& GetRanges()
{ return fRanges; }
std::vector<TMVA::GeneticGenes>& GetGenePool()
{ return fGenePool; }
std::vector<TMVA::GeneticRange*>& GetRanges()
{ return fRanges; }
void NextGeneration()
{}