OnlineLibLinear.cpp

Go to the documentation of this file.

/*
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation; either version 3 of the License, or
 * (at your option) any later version.
 *
 * Written (W) 2007-2010 Soeren Sonnenburg
 * Written (W) 2011 Shashwat Lal Das
 * Copyright (c) 2007-2009 The LIBLINEAR Project.
 * Copyright (C) 2007-2010 Fraunhofer Institute FIRST and Max-Planck-Society
 */

#include <shogun/classifier/svm/OnlineLibLinear.h>

using namespace shogun;

COnlineLibLinear::COnlineLibLinear()
        : COnlineLinearMachine()
{
        init();
}

COnlineLibLinear::COnlineLibLinear(float64_t C)
{
        init();
        C1=C;
        C2=C;
        use_bias=true;
}

COnlineLibLinear::COnlineLibLinear(
        float64_t C, CStreamingDotFeatures* traindat)
{
        init();
        C1=C;
        C2=C;
        use_bias=true;

        set_features(traindat);
}


void COnlineLibLinear::init()
{
        C1=1;
        C2=1;
        use_bias=false;

        m_parameters->add(&C1, "C1",  "C Cost constant 1.");
        m_parameters->add(&C2, "C2",  "C Cost constant 2.");
        m_parameters->add(&use_bias, "use_bias",  "Indicates if bias is used.");
}

COnlineLibLinear::~COnlineLibLinear()
{
}

bool COnlineLibLinear::train(CFeatures* data)
{
        if (data)
        {
                if (!data->has_property(FP_STREAMING_DOT))
                        SG_ERROR("Specified features are not of type CStreamingDotFeatures\n");
                set_features((CStreamingDotFeatures*) data);
        }
        
        float64_t C, d, G;
        float64_t QD;

        // y and alpha for example being processed
        int32_t y_current;
        float64_t alpha_current;

        // Cost constants
        float64_t Cp=C1;
        float64_t Cn=C2;

        // PG: projected gradient, for shrinking and stopping
        float64_t PG;
        float64_t PGmax_old = CMath::INFTY;
        float64_t PGmin_old = -CMath::INFTY;
        float64_t PGmax_new = -CMath::INFTY;
        float64_t PGmin_new = CMath::INFTY;

        // Diag is probably unnecessary
        float64_t diag[3] = {0, 0, 0};
        float64_t upper_bound[3] = {Cn, 0, Cp};

        // Bias
        bias = 0;

        PGmax_new = -CMath::INFTY;
        PGmin_new = CMath::INFTY;

        // Objective value = v/2
        float64_t v = 0;
        // Number of support vectors
        int32_t nSV = 0;

        // Start reading the examples
        features->start_parser();

        CTime start_time;
        while (features->get_next_example())
        {
                alpha_current = 0;
                if (features->get_label() > 0)
                        y_current = +1;
                else
                        y_current = -1;

                QD = diag[y_current + 1];
                // Dot product of vector with itself
                QD += features->dot(features);

                features->expand_if_required(w, w_dim);

                G = features->dense_dot(w, w_dim);
                if (use_bias)
                        G += bias;
                G = G*y_current - 1;
                // LINEAR TERM PART?

                C = upper_bound[y_current + 1];
                G += alpha_current*diag[y_current + 1]; // Can be eliminated, since diag = 0 vector

                PG = 0;
                if (alpha_current == 0) // This condition will always be true in the online version
                {
                        if (G > PGmax_old)
                        {
                                features->release_example();
                                continue;
                        }
                        else if (G < 0)
                                PG = G;
                }
                else if (alpha_current == C)
                {
                        if (G < PGmin_old)
                        {
                                features->release_example();
                                continue;
                        }
                        else if (G > 0)
                                PG = G;
                }
                else
                        PG = G;

                PGmax_new = CMath::max(PGmax_new, PG);
                PGmin_new = CMath::min(PGmin_new, PG);

                if (fabs(PG) > 1.0e-12)
                {
                        float64_t alpha_old = alpha_current;
                        alpha_current = CMath::min(CMath::max(alpha_current - G/QD, 0.0), C);
                        d = (alpha_current - alpha_old) * y_current;

                        features->add_to_dense_vec(d, w, w_dim);

                        if (use_bias)
                                bias += d;
                }

                v += alpha_current*(alpha_current*diag[y_current + 1] - 2);
                if (alpha_current > 0)
                        nSV++;

                features->release_example();
        }

        features->end_parser();

        float64_t gap = PGmax_new - PGmin_new;

        SG_DONE();
        SG_INFO("Optimization finished.\n");

        // calculate objective value
        for (int32_t i=0; i<w_dim; i++)
                v += w[i]*w[i];
        v += bias*bias;

        SG_INFO("Objective value = %lf\n", v/2);
        SG_INFO("nSV = %d\n", nSV);
        SG_INFO("gap = %g\n", gap);

        return true;
}