Logo ROOT  
Reference Guide
TMVAGAexample2.C
Go to the documentation of this file.
1/// \file
2/// \ingroup tutorial_tmva
3/// \notebook -nodraw
4/// This executable gives an example of a very simple use of the genetic algorithm
5/// of TMVA.
6/// - Project : TMVA - a Root-integrated toolkit for multivariate data analysis
7/// - Package : TMVA
8/// - Executable: TMVAGAexample
9///
10/// \macro_output
11/// \macro_code
12/// \author Andreas Hoecker
13
14
15#include <iostream> // Stream declarations
16#include <vector>
17
19#include "TMVA/GeneticFitter.h"
20#include "TMVA/IFitterTarget.h"
21
22using namespace std;
23
24namespace TMVA {
25
26
27class MyFitness : public IFitterTarget {
28 public:
29 MyFitness() : IFitterTarget() {
30 }
31
32 // the fitness-function goes here
33 // the factors are optimized such that the return-value of this function is minimized
34 // take care!! the fitness-function must never fail, .. means: you have to prevent
35 // the function from reaching undefined values (such as x=0 for 1/x or so)
36 //
37 // HINT: to use INTEGER variables, it is sufficient to cast the "factor" in the fitness-function
38 // to (int). In this case the variable-range has to be chosen +1 ( to get 0..5, take Interval(0,6) )
39 // since the introduction of "Interval" ranges can be defined with a third parameter
40 // which gives the number of bins within the interval. With that technique discrete values
41 // can be achieved easier. The random selection out of this discrete numbers is completely uniform.
42 //
43 Double_t EstimatorFunction( std::vector<Double_t> & factors ){
44 //return (10.- (int)factors.at(0) *factors.at(1) + (int)factors.at(2));
45 return (10.- factors.at(0) *factors.at(1) + factors.at(2));
46
47 //return 100.- (10 + factors.at(1)) *factors.at(2)* TMath::Abs( TMath::Sin(factors.at(0)) );
48 }
49};
50
51
52
53
54
55
56
57
58void exampleGA(){
59 std::cout << "\nEXAMPLE" << std::endl;
60 // define all the parameters by their minimum and maximum value
61 // in this example 3 parameters are defined.
62 vector<Interval*> ranges;
63 ranges.push_back( new Interval(0,15,30) );
64 ranges.push_back( new Interval(0,13) );
65 ranges.push_back( new Interval(0,5,3) );
66
67 for( std::vector<Interval*>::iterator it = ranges.begin(); it != ranges.end(); it++ ){
68 std::cout << " range: " << (*it)->GetMin() << " " << (*it)->GetMax() << std::endl;
69 }
70
71 IFitterTarget* myFitness = new MyFitness();
72
73 // prepare the genetic algorithm with an initial population size of 20
74 // mind: big population sizes will help in searching the domain space of the solution
75 // but you have to weight this out to the number of generations
76 // the extreme case of 1 generation and populationsize n is equal to
77 // a Monte Carlo calculation with n tries
78
79 const TString name( "exampleGA" );
80 const TString opts( "PopSize=100:Steps=30" );
81
82 GeneticFitter mg( *myFitness, name, ranges, opts);
83 // mg.SetParameters( 4, 30, 200, 10,5, 0.95, 0.001 );
84
85 std::vector<Double_t> result;
86 Double_t estimator = mg.Run(result);
87
88 int n = 0;
89 for( std::vector<Double_t>::iterator it = result.begin(); it<result.end(); it++ ){
90 std::cout << "FACTOR " << n << " : " << (*it) << std::endl;
91 n++;
92 }
93
94}
95
96
97
98} // namespace TMVA
99
100void TMVAGAexample2() {
101 cout << "Start Test TMVAGAexample" << endl
102 << "========================" << endl
103 << endl;
104
105 TMVA::exampleGA();
106
107}
108
109
110int main( int argc, char** argv )
111{
112 TMVAGAexample2();
113 return 0;
114}
double Double_t
Definition: RtypesCore.h:59
char name[80]
Definition: TGX11.cxx:110
int main(int argc, char *argv[])
Definition: cef_main.cxx:54
Basic string class.
Definition: TString.h:136
const Int_t n
Definition: legend1.C:16
static constexpr double mg
create variable transformations