example_perceptron_kernel_gaussian_gmm.cpp

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

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

Para executar o programa:

./example_perceptron_kernel_gaussian_gmm

Retornando por consola:

Neurona:
-0.084801134424809 6.5644266025979    6.7976090202446    7.2466923286193    7.0344311122582    6.4082894628588    6.0603114182925    6.4848189537356    8.6777704375106    7.5236535836757    7.4045759280255    -6.7251099705574   -7.6567594525611   -8.6558523629629   -6.9612035074631   -7.4106021940437   -5.9315724812651   -6.3772318212795   -6.470017398032    -5.1911400757936   -7.1811853152788   

╔═══════════════════════════════════╗
║    ClassificationMetrics data     ║
╠═══════════════════════════════════╣
║        Threshold: 0.5             ║
║          Samples: 2000            ║
╠═══════════════════════════════════╣
║ Pred.[0]/Act.[*]: 974     17      ║
║ Pred.[1]/Act.[*]: 26      983     ║
╠═══════════════════════════════════╣
║         Accuracy: 97.85 %         ║
║        Precision: 97.42 %         ║
║           Recall: 98.3  %         ║
╠═══════════════════════════════════╣
║           FScore: 97.86 %         ║
╚═══════════════════════════════════╝

╔═══════════════════════════════════╗
║    ClassificationMetrics data     ║
╠═══════════════════════════════════╣
║        Threshold: 0.5             ║
║          Samples: 2000            ║
╠═══════════════════════════════════╣
║ Pred.[0]/Act.[*]: 959     25      ║
║ Pred.[1]/Act.[*]: 41      975     ║
╠═══════════════════════════════════╣
║         Accuracy: 96.7  %         ║
║        Precision: 95.96 %         ║
║           Recall: 97.5  %         ║
╠═══════════════════════════════════╣
║           FScore: 96.73 %         ║
╚═══════════════════════════════════╝

Classification - data

Classification - training

Classification - testing

Código example_perceptron_kernel_gaussian_gmm.cpp:

 
#include <iostream>
#include <Pds/Ra>
#include <Pds/Ml>
 
    
int main(void)
{
    Pds::ClassificationMetrics Metrics;
    Pds::Vector Yp;
    Pds::Matrix F;
    Pds::IterationConf Conf;
    Conf.SetMaxIter(10000); Conf.Show=false;
    unsigned int K=10;
    
    // Generating data
    unsigned int L=1000;
    Pds::Matrix X;
    Pds::Vector Y;
    Pds::Matrix G;
    Pds::DataSet::Classification::LoadDataYinYang(L,X,Y);
    
    Pds::Octave::XLabel="x_1";
    Pds::Octave::YLabel="x_2";
    Pds::Octave::Plot::PointsX2DY(X,Y,"testando.m","example_perceptron_kernel_gaussian_gmm_data.png");
    
    std::vector<Pds::Matrix> Block=X.FindRowsBlock(Y);
    Pds::DataGMM Dat0=Pds::Clustering::GMM(Conf,Block[0],K,1.0);
    Pds::DataGMM Dat1=Pds::Clustering::GMM(Conf,Block[1],K,1.0);
    Dat1.Mu.insert  (Dat1.Mu.end()  , Dat0.Mu.begin()  , Dat0.Mu.end());
    Dat1.Sinv.insert(Dat1.Sinv.end(), Dat0.Sinv.begin(), Dat0.Sinv.end());
    
    // Create Perceptron
    F=Pds::Kernel::Gaussian(X,Dat1.Mu,Dat1.Sinv);
 
    Pds::Perceptron Neurona(Conf,F,Y);
    Neurona.Print("\nNeurona:\n");
    
    // Predict training data
    Yp=Neurona.Predict(F);
    
    // Metrics of training
    Metrics = Pds::ClassificationMetrics::Calculate(0.5,Yp,Y);
    Metrics.Print("\n");
    
    Pds::Octave::Plot::PointsX2DY(X,Yp,"testando.m","example_perceptron_kernel_gaussian_gmm_training.png");
        
    // Predict testing data
    Pds::DataSet::Classification::LoadDataYinYang(L,X,Y); // Load new data
    
    F=Pds::Kernel::Gaussian(X,Dat1.Mu,Dat1.Sinv);
    Yp=Neurona.Predict(F);
    
    // Metrics testing
    Metrics = Pds::ClassificationMetrics::Calculate(Metrics.Threshold,Yp,Y);
    Metrics.Print("\n");
    
    Pds::Octave::Plot::PointsX2DY(X,Yp,"testando.m","example_perceptron_kernel_gaussian_gmm_testing.png");
    return 0;
}