example_logisticregression_ig2.cpp

Programa para el testeo de las funciones.Para compilar o código example_logisticregression_ig2.cpp:

g++ -static -o example_logisticregression_ig2 example_logisticregression_ig2.cpp -lpdsmlmm -lpdsramm -lpdsspmm

Para executar o programa:

./example_logisticregression_ig2

Retornando por consola:

{X,Y}
2    6    1500 22   0    
4    6    1500 23   1    
2    6    1500 28   1    
2    6    1500 28   0    
4    6    1500 35   0    
4    6    1500 39   0    
2    6    1500 22   1    
2    6    1500 26   0    
2    6    1500 26   1    
2    6    1500 28   1    
2    6    1500 35   1    
4    6    1500 28   1    
4    6    1500 30   0    
4    6    1500 35   0    
4    6    1500 35   1    
4    6    1500 38   1    

    W_ig: 4.6454166929886   0                 0                 0                 -0.12066017384386 
 Cost(W): 0.078433225987246

    W_ig: 251.21391246223   -0.16324517096075 -0.16324517096075 -0.16324517096075 -0.16324517096075 
 Cost(W): 0.041120692117862

Código example_logisticregression_ig2.cpp:

 
#include <iostream>
#include <Pds/Ra>
#include <Pds/Ml>
        
int main(void)
{
    Pds::IterationConf Conf;    
    Conf.Show=false; Conf.SetMinError(1e-07);    Conf.SetAlpha(0.1);
    
    // Generating data
    Pds::Vector Y="0    1       1       0       0       0       1       0       1       1       1       1       0       0       1       1";
 
    Pds::Matrix X=
    "2  4       2       2       4       4       2       2       2       2       2       4       4       4       4       4\n"
    "6  6       6       6       6       6       6       6       6       6       6       6       6       6       6       6\n"
    "1500       1500    1500    1500    1500    1500    1500    1500    1500    1500    1500    1500    1500    1500    1500    1500\n"
    "22 23      28      28      35      39      22      26      26      28      35      28      30      35      35      38";
 
    X=X.T();
    Y=Y.T();
    
    Pds::MergeHor({X,Y}).Print("{X,Y}\n");
    
    Pds::Vector Yp;
    Pds::Vector W;
    
    std::cout<<"\n\n";    
    // Create W using training data
    
    // Método ortogonal
    W=Pds::LogisticModel::FittingOrtogonalIG(X,Y,1);
 
    W.T().Print("    W_ig: ");
    std::cout<< " Cost(W): "<<Pds::LogisticModel::CostInformationGain(W,X,Y)<<"\n\n";    
    
    Yp=Pds::LogisticModel::Classify(W,X);
    
    // Método de gradiente
    //W=Pds::LogisticModel::GetW0MeanMethod(X);   
    W=Pds::LogisticModel::FittingGradientIG(Conf,X,Y,W);
    
    W.T().Print("    W_ig: ");
    std::cout<< " Cost(W): "<<Pds::LogisticModel::CostInformationGain(W,X,Y)<<"\n"; 
    
    Yp=Pds::LogisticModel::Classify(W,X);
    
    return 0;
}