LibLinearMTL.h

浏览该文件的文档.

/*
 * 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) 2011-2012 Christian Widmer
 * Written (W) 2007-2010 Soeren Sonnenburg
 * Copyright (c) 2007-2009 The LIBLINEAR Project.
 * Copyright (C) 2007-2012 Fraunhofer Institute FIRST and Max-Planck-Society
 */

#ifndef _LIBLINEARMTL_H___
#define _LIBLINEARMTL_H___

#include <shogun/lib/config.h>

#include <shogun/lib/common.h>
#include <shogun/base/Parameter.h>
#include <shogun/machine/LinearMachine.h>
#include <shogun/optimization/liblinear/shogun_liblinear.h>
#include <shogun/lib/SGSparseMatrix.h>
#include <shogun/lib/SGSparseVector.h>

#include <map>

namespace shogun
{

#ifdef HAVE_LAPACK


class MappedSparseMatrix
{

    public:

    inline const float64_t operator()(index_t i_row, index_t i_col) const
    {

        // lookup complexity is O(log n)
        std::map<index_t, float64_t>::const_iterator it = data[i_row].find(i_col);

        if (it != data[i_row].end())
        {
            // use mapping for lookup
            return it->second;
        } else {
            return 0.0;
        }
    }

    void set_from_sparse(const SGSparseMatrix<float64_t> &sgm)
    {
        data.clear();

        // deep copy sparse matrix
        for (int32_t i=0; i!=sgm.num_vectors; i++)
        {

            SGSparseVector<float64_t> ts_row = sgm.sparse_matrix[i];
            data.push_back(std::map<index_t, float64_t>());

            for (int32_t k=0; k!=ts_row.num_feat_entries; k++)
            {
                // get data from sparse matrix
                SGSparseVectorEntry<float64_t> e = ts_row.features[k];
                data[i][e.feat_index] = e.entry;
            }

        }
    }

    std::vector< std::map<index_t, float64_t> > data;

};


class CLibLinearMTL : public CLinearMachine
{
    public:
        CLibLinearMTL();


        CLibLinearMTL(
            float64_t C, CDotFeatures* traindat,
            CLabels* trainlab);

        virtual ~CLibLinearMTL();


        virtual EMachineType get_classifier_type() { return CT_LIBLINEAR; }

        inline void set_C(float64_t c_neg, float64_t c_pos) { C1=c_neg; C2=c_pos; }

        inline float64_t get_C1() { return C1; }

        inline float64_t get_C2() { return C2; }

        inline void set_epsilon(float64_t eps) { epsilon=eps; }

        inline float64_t get_epsilon() { return epsilon; }

        inline void set_bias_enabled(bool enable_bias) { use_bias=enable_bias; }

        inline bool get_bias_enabled() { return use_bias; }

        virtual const char* get_name() const { return "LibLinearMTL"; }

        inline int32_t get_max_iterations()
        {
            return max_iterations;
        }

        inline void set_max_iterations(int32_t max_iter=1000)
        {
            max_iterations=max_iter;
        }

        inline void set_num_tasks(int32_t nt)
        {
            num_tasks = nt;
        }

        inline void set_linear_term(SGVector<float64_t> linear_term)
        {
            if (!m_labels)
                SG_ERROR("Please assign labels first!\n")

            int32_t num_labels=m_labels->get_num_labels();

            if (num_labels!=linear_term.vlen)
            {
                SG_ERROR("Number of labels (%d) does not match number"
                        " of entries (%d) in linear term \n", num_labels,
                        linear_term.vlen);
            }

            m_linear_term = linear_term;
        }

        inline void set_task_indicator_lhs(SGVector<int32_t> ti)
        {
            task_indicator_lhs = ti;
        }

        inline void set_task_indicator_rhs(SGVector<int32_t> ti)
        {
            task_indicator_rhs = ti;
        }

        inline void set_task_similarity_matrix(SGSparseMatrix<float64_t> tsm)
        {
            task_similarity_matrix.set_from_sparse(tsm);
        }

        inline void set_graph_laplacian(SGMatrix<float64_t> lap)
        {
            graph_laplacian = lap;
        }

        inline SGMatrix<float64_t> get_V()
        {
            return V;
        }

        inline SGMatrix<float64_t> get_W()
        {

            int32_t w_size = V.num_rows;

            SGMatrix<float64_t> W = SGMatrix<float64_t>(w_size, num_tasks);
            for(int32_t k=0; k<w_size*num_tasks; k++)
            {
                W.matrix[k] = 0;
            }

            for (int32_t s=0; s<num_tasks; s++)
            {
                float64_t* v_s = V.get_column_vector(s);
                for (int32_t t=0; t<num_tasks; t++)
                {
                    float64_t sim_ts = task_similarity_matrix(s,t);
                    for(int32_t i=0; i<w_size; i++)
                    {
                        W.matrix[t*w_size + i] += sim_ts * v_s[i];
                    }
                }
            }

            return W;
        }

        inline SGVector<float64_t> get_alphas()
        {
            return alphas;
        }

        virtual float64_t compute_primal_obj();

        virtual float64_t compute_dual_obj();

        virtual float64_t compute_duality_gap();


    protected:
        virtual bool train_machine(CFeatures* data=NULL);

    private:
        void init();

        void solve_l2r_l1l2_svc(
            const liblinear_problem *prob, double eps, double Cp, double Cn);


    protected:
        float64_t C1;
        float64_t C2;
        bool use_bias;
        float64_t epsilon;
        int32_t max_iterations;

        SGVector<float64_t> m_linear_term;

        SGVector<float64_t> alphas;

        int32_t num_tasks;

        SGVector<int32_t> task_indicator_lhs;

        SGVector<int32_t> task_indicator_rhs;

        //SGMatrix<float64_t> task_similarity_matrix;
        //SGSparseMatrix<float64_t> task_similarity_matrix;
        MappedSparseMatrix task_similarity_matrix;

        SGMatrix<float64_t> graph_laplacian;

        SGMatrix<float64_t> V;

        float64_t duality_gap;

};

#endif //HAVE_LAPACK

} /* namespace shogun  */

#endif //_LIBLINEARMTL_H___