example_perceptron_kernel_gaussian_yinyang.cpp

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

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

Para executar o programa:

./example_perceptron_kernel_gaussian_yinyang

Retornando por consola:

┌──────────────────────────────┐
│    Kmeans                    │
│    IterationConf init data   │
├──────────────────────────────┤
│        MinError: 1.0000e-05  │
│         MaxIter: 1e+04       │
└──────────────────────────────┘
┌──────────────────────────────┐
│            Iter: 22          │
│       LastError: 0.0000e+00  │
│ Elapsed time(s): 0.00029     │
└──────────────────────────────┘
┌──────────────────────────────┐
│    Kmeans                    │
│    IterationConf init data   │
├──────────────────────────────┤
│        MinError: 1.0000e-05  │
│         MaxIter: 1e+04       │
└──────────────────────────────┘
┌──────────────────────────────┐
│            Iter: 14          │
│       LastError: 0.0000e+00  │
│ Elapsed time(s): 0.000545    │
└──────────────────────────────┘

Neurona:
-0.5636647240059        4.2500671732378 5.4892010147833 5.3230443694196 -0.79303785676519       3.3063046749552 3.3613114987862 2.1955379034242 2.938246573117  1.8896910410267 0.085686982708166       3.7274552317887 5.7643894904731 3.3514813420701 -0.44825547327836       4.5324916859488 4.9867742065307 1.7749574275709 5.2433832244008 4.0659870664641 4.166780241898  2.2914392339366 5.45212789566411.0044276383258  4.4705395627767 2.3167806979269 4.3070381347576 3.4792946126242 4.2731504272416 3.6220403005052 4.0473395344371 5.7552191007249 3.0955220962961 2.9859493831289 2.4707323903569 0.55892896589753        4.9660083796706 0.95822437302003        4.4011969499203 2.8034196827806 4.5165444193199 4.0146951120763 2.6972829247613 3.8529558751443 1.1776467889837 4.1031636581869 2.05026783008355.6837202333931  0.6971254244837 2.1468060223755 8.1671378554394 3.1121086575305 3.356313090809  4.1925353817637 2.5714048833768 4.1664511072377 4.7111517197842 0.52444367568057        -1.7028097821304        1.5570539154892 4.5261158948409 -5.8551030502859        5.1537207802395 -1.564392905755 -1.0164655332193        -2.4341794874837        -4.0647004118528        -1.2719442894787        -3.6241326267765        -5.9005867516651        -1.9020763964996        -3.1480734109237        -4.085845436565 -2.2230498257697        -1.9348291596943        -4.9716760750194        -0.82372734825709       -6.7954870541113        -5.5916737524146        -4.1626808981897        -3.3620991776216        -1.2849891036665        -4.801136088596 -4.1978443857668        -3.3848607420899        -4.4170976696187        -5.3138657579617        0.75099192767982        -5.7527521039533        -2.7576193948384        0.12884007638803        -2.9539921941337        -3.5603954216387        -3.6908682066319        -2.168710562259-2.0559827864865 -1.3752807234752        -2.5769046038249        -1.4491630891129        -1.0682296915817        -4.7770862613907        -4.0243210525499        0.10255667798547        -0.46089061402273       -2.5046122271292        -1.9215637190145        -2.8572235393748        -1.4897494416152        -1.9977801075796        -1.7174822419324        -0.56322753832583       -3.1437621583034        -5.993908810349 -0.089989509639045      -4.7789178035185        -3.1564677923807        -5.9983397118331        -3.2812792656054        -3.0849490108091        -1.2786220304571        0.45286188790035

╔═══════════════════════════════════╗
║    ClassificationMetrics data     ║
╠═══════════════════════════════════╣
║        Threshold: 0.5             ║
║          Samples: 2000            ║
╠═══════════════════════════════════╣
║ Pred.[0]/Act.[*]: 994     17      ║
║ Pred.[1]/Act.[*]: 6       983     ║
╠═══════════════════════════════════╣
║         Accuracy: 98.85 %         ║
║        Precision: 99.39 %         ║
║           Recall: 98.3  %         ║
╠═══════════════════════════════════╣
║           FScore: 98.84 %         ║
╚═══════════════════════════════════╝

╔═══════════════════════════════════╗
║    ClassificationMetrics data     ║
╠═══════════════════════════════════╣
║        Threshold: 0.5             ║
║          Samples: 2000            ║
╠═══════════════════════════════════╣
║ Pred.[0]/Act.[*]: 988     23      ║
║ Pred.[1]/Act.[*]: 12      977     ║
╠═══════════════════════════════════╣
║         Accuracy: 98.25 %         ║
║        Precision: 98.79 %         ║
║           Recall: 97.7  %         ║
╠═══════════════════════════════════╣
║           FScore: 98.24 %         ║
╚═══════════════════════════════════╝

Classification - data

Classification - training

Classification - testing

Código example_perceptron_kernel_gaussian_yinyang.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;
    double Sigma=0.1;
    
    // 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_yinyang_data.png");
    
    // Create Perceptron
    G=Pds::MergeVer( {Pds::Clustering::Kmeans(Conf,X.GetRows(Pds::Find(Y)),60),
                      Pds::Clustering::Kmeans(Conf,X.GetRows(Pds::Find(1-Y)), 60)}
                   ); //Random centroids to Gaussian Kernel
    F=Pds::Kernel::Gaussian(X,G,Sigma);
 
    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_yinyang_training.png");
        
    // Predict testing data
    Pds::DataSet::Classification::LoadDataYinYang(L,X,Y); // Load new data
    
    F=Pds::Kernel::Gaussian(X,G,Sigma);
    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_yinyang_testing.png");
    return 0;
}