example_perceptron_evaluate.cpp

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

g++ -static -o example_perceptron_evaluate example_perceptron_evaluate.cpp -lpdsnnmm 

Para executar o programa:

./example_perceptron_evaluate

Retornando por consola:

GradientCE:
Error: RMS(△ W)/Std(W)
-------------------------
   IterationConf data    
-------------------------
    Alpha: 0.1
   Lambda: 0.001
 MinError: 1e-05
  MaxIter: 10000
-------------------------
LastError: 4.1538e-05
 LastIter: 10000
-------------------------
    
Neurona:
0.064281158965802       10.819510055137 10.96149408058
    
-----------------------------
 ClassificationMetrics data  
-----------------------------
Threshold       : 0.52
Samples         : 200
----------------:------------
Pred.[0]/Act.[*]: 100   0
Pred.[1]/Act.[*]: 0     100
----------------:------------
Accuracy        : 100%
Precision       : 100%
Recall          : 100%
----------------:------------
FScore          : 100%
-----------------------------
    
-----------------------------
 ClassificationMetrics data  
-----------------------------
Threshold       : 0.52
Samples         : 200
----------------:------------
Pred.[0]/Act.[*]: 100   1
Pred.[1]/Act.[*]: 0     99
----------------:------------
Accuracy        : 99.5%
Precision       : 100%
Recall          : 99%
----------------:------------
FScore          : 99.5%
-----------------------------

Código example_perceptron_evaluate.cpp:

 
#include <iostream>
#include <Pds/Ra>
#include <Pds/Ml>
 
    
int main(void)
{
    Pds::IterationConf Conf; Conf.SetMaxIter(1000); Conf.Show=true;
    Pds::ClassificationMetrics Metrics;
    Pds::Vector Yp;
    
    // Generating data
    unsigned int L=100000;
    Pds::Matrix X;
    Pds::Vector Y;
    Pds::DataSet::Classification::LoadDataLine(L,X,Y);
 
    // Split data set in {Training,Cross-validation,Test}
    Pds::DataSetBlock Dat=Pds::DataSet::Split(X,Y,6,2,2);
    
    // Create Perceptron
    Pds::Ra::Tic();
    Pds::Perceptron Neurona(Conf,Dat.Xtr,Dat.Ytr);
    Pds::Ra::Toc();
    Neurona.Print("\nNeurona:\n");
    
    // Evaluate training
    Yp=Neurona.Evaluate(Dat.Xtr);
    Metrics = Pds::ClassificationMetrics::Calculate(0.5,Yp,Dat.Ytr);
    Metrics.Print("\nTraining:\n");
 
    // Evaluate cross-validation
    Yp=Neurona.Evaluate(Dat.Xcv);
    Pds::Ra::Tic();
    Metrics = Pds::ClassificationMetrics::Optimal(Yp,Dat.Ycv);
    Metrics.Print("\nCross-validation:\n");
    Pds::Ra::Toc();
    
    // Evaluate testing data
    Yp=Neurona.Evaluate(Dat.Xtt);
    Metrics = Pds::ClassificationMetrics::Calculate(Metrics.Threshold,Yp,Dat.Ytt);
    Metrics.Print("\nTesting:\n");
    
    return 0;
}