logistic_regression_LBFGS.cpp

Go to the documentation of this file.

#include <shark/Data/SparseData.h>
#include <shark/ObjectiveFunctions/Loss/CrossEntropy.h>
#include <shark/ObjectiveFunctions/Regularizer.h>
 
#include <shark/Algorithms/GradientDescent/LBFGS.h>
#include <shark/ObjectiveFunctions/ErrorFunction.h>
#include <shark/Models/LinearModel.h>
 
#include <shark/Core/Timer.h>
#include <iostream>
using namespace shark;
using namespace std;
 
int main(int argc, char **argv) {
    ClassificationDataset data;
    importSparseData(data, "mnist",0,8192);
    double alpha = 0.1;
    CrossEntropy<unsigned int, RealVector> loss;
    LinearClassifier<> model;
    
    //Setting up the problem
    model.decisionFunction().setStructure(inputDimension(data),numberOfClasses(data),true);
    TwoNormRegularizer regularizer;
    ErrorFunction error(data,&model.decisionFunction(),&loss);
    error.setRegularizer(alpha,&regularizer);
    
    //solving
    Timer time;
    LBFGS optimizer;
    optimizer.init(error);
    while(error.evaluationCounter()<200){
        optimizer.step(error);
    }
    model.setParameterVector(optimizer.solution().point);
    double time_taken = time.stop();
    
    cout << "Cross-Entropy: " << loss(data.labels(),model.decisionFunction()(data.inputs()))<<std::endl;
    cout << "Time:\n" << time_taken << endl;
}