SalzbergWordStringKernel.h

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) 1999-2008 Gunnar Raetsch
 * Copyright (C) 1999-2009 Fraunhofer Institute FIRST and Max-Planck-Society
 */

#ifndef _SALZBERGWORDSTRINGKERNEL_H___
#define _SALZBERGWORDSTRINGKERNEL_H___

#include <shogun/lib/config.h>

#include <shogun/lib/common.h>
#include <shogun/kernel/string/StringKernel.h>
#include <shogun/classifier/PluginEstimate.h>
#include <shogun/features/StringFeatures.h>

namespace shogun
{
class CSalzbergWordStringKernel: public CStringKernel<uint16_t>
{
    public:
        CSalzbergWordStringKernel();

        CSalzbergWordStringKernel(int32_t size, CPluginEstimate* pie, CLabels* labels=NULL);

        CSalzbergWordStringKernel(
            CStringFeatures<uint16_t>* l, CStringFeatures<uint16_t>* r,
            CPluginEstimate *pie, CLabels* labels=NULL);

        virtual ~CSalzbergWordStringKernel();

        void set_prior_probs(float64_t pos_prior_, float64_t neg_prior_)
        {
            pos_prior=pos_prior_ ;
            neg_prior=neg_prior_ ;
            if (fabs(pos_prior+neg_prior-1)>1e-6)
                SG_WARNING("priors don't sum to 1: %f+%f-1=%f\n", pos_prior, neg_prior, pos_prior+neg_prior-1)
        };

        void set_prior_probs_from_labels(CLabels* labels);

        virtual bool init(CFeatures* l, CFeatures* r);

        virtual void cleanup();

        virtual EKernelType get_kernel_type() { return K_SALZBERG; }

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

    protected:
        float64_t compute(int32_t idx_a, int32_t idx_b);
        //  float64_t compute_slow(int64_t idx_a, int64_t idx_b);

        inline int32_t compute_index(int32_t position, uint16_t symbol)
        {
            return position*num_symbols+symbol;
        }
    private:
        void init();

    protected:
        CPluginEstimate* estimate;

        float64_t* mean;
        float64_t* variance;

        float64_t* sqrtdiag_lhs;
        float64_t* sqrtdiag_rhs;

        float64_t* ld_mean_lhs;
        float64_t* ld_mean_rhs;

        int32_t num_params;
        int32_t num_symbols;
        float64_t sum_m2_s2;
        float64_t pos_prior;
        float64_t neg_prior;
        bool initialized;
};
}
#endif /* _SALZBERGWORDKERNEL_H__ */