TanimotoKernelNormalizer.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) 2008-2009 Soeren Sonnenburg
 * Copyright (C) 2008-2009 Fraunhofer Institute FIRST and Max-Planck-Society
 */

#ifndef _TANIMOTOKERNELNORMALIZER_H___
#define _TANIMOTOKERNELNORMALIZER_H___

#include <shogun/lib/config.h>

#include <shogun/kernel/normalizer/KernelNormalizer.h>
#include <shogun/kernel/string/CommWordStringKernel.h>

namespace shogun
{
class CTanimotoKernelNormalizer : public CKernelNormalizer
{
    public:
        CTanimotoKernelNormalizer(bool use_opt_diag=false)
            : CKernelNormalizer(), diag_lhs(NULL), diag_rhs(NULL),
            use_optimized_diagonal_computation(use_opt_diag)
        {
        }

        virtual ~CTanimotoKernelNormalizer()
        {
            SG_FREE(diag_lhs);
            SG_FREE(diag_rhs);
        }

        virtual bool init(CKernel* k)
        {
            ASSERT(k)
            int32_t num_lhs=k->get_num_vec_lhs();
            int32_t num_rhs=k->get_num_vec_rhs();
            ASSERT(num_lhs>0)
            ASSERT(num_rhs>0)

            CFeatures* old_lhs=k->lhs;
            CFeatures* old_rhs=k->rhs;

            k->lhs=old_lhs;
            k->rhs=old_lhs;
            bool r1=alloc_and_compute_diag(k, diag_lhs, num_lhs);

            k->lhs=old_rhs;
            k->rhs=old_rhs;
            bool r2=alloc_and_compute_diag(k, diag_rhs, num_rhs);

            k->lhs=old_lhs;
            k->rhs=old_rhs;

            return r1 && r2;
        }

        virtual float64_t normalize(
            float64_t value, int32_t idx_lhs, int32_t idx_rhs)
        {
            float64_t diag_sum=diag_lhs[idx_lhs]*diag_rhs[idx_rhs];
            return value/(diag_sum-value);
        }

        virtual float64_t normalize_lhs(float64_t value, int32_t idx_lhs)
        {
            SG_ERROR("linadd not supported with Tanimoto normalization.\n")
            return 0;
        }

        virtual float64_t normalize_rhs(float64_t value, int32_t idx_rhs)
        {
            SG_ERROR("linadd not supported with Tanimoto normalization.\n")
            return 0;
        }

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

    public:
        bool alloc_and_compute_diag(CKernel* k, float64_t* &v, int32_t num)
        {
            SG_FREE(v);
            v=SG_MALLOC(float64_t, num);

            for (int32_t i=0; i<num; i++)
            {
                if (k->get_kernel_type() == K_COMMWORDSTRING)
                {
                    if (use_optimized_diagonal_computation)
                        v[i]=((CCommWordStringKernel*) k)->compute_diag(i);
                    else
                        v[i]=((CCommWordStringKernel*) k)->compute_helper(i,i, true);
                }
                else
                    v[i]=k->compute(i,i);

                if (v[i]==0.0)
                    v[i]=1e-16; /* avoid divide by zero exception */
            }

            return (v!=NULL);
        }

    protected:
        float64_t* diag_lhs;
        float64_t* diag_rhs;
        bool use_optimized_diagonal_computation;
};
}
#endif