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

#ifndef _EXPLICITSPECFEATURES_H___
#define _EXPLICITSPECFEATURES_H___

#include "lib/common.h"
#include "lib/io.h"
#include "features/DotFeatures.h"
#include "features/StringFeatures.h"

namespace shogun
{
template <class ST> class CStringFeatures;

class CExplicitSpecFeatures : public CDotFeatures
{
    public:
        CExplicitSpecFeatures(void);

        CExplicitSpecFeatures(CStringFeatures<uint16_t>* str, bool normalize=true);

        CExplicitSpecFeatures(const CExplicitSpecFeatures & orig);

        virtual ~CExplicitSpecFeatures();

        virtual CFeatures* duplicate() const;

        inline virtual int32_t get_dim_feature_space()
        {
            return spec_size;
        }

        virtual float64_t dot(int32_t vec_idx1, CDotFeatures* df, int32_t vec_idx2);

        virtual float64_t dense_dot(int32_t vec_idx1, const float64_t* vec2, int32_t vec2_len);

        virtual void add_to_dense_vec(float64_t alpha, int32_t vec_idx1, float64_t* vec2, int32_t vec2_len, bool abs_val=false);

        struct explicit_spec_feature_iterator
        {
            uint16_t* vec;
            int32_t vidx;
            int32_t vlen;
            bool vfree;

            int32_t index;

        };

        virtual void* get_feature_iterator(int32_t vector_index)
        {
            SG_NOTIMPLEMENTED;
            return NULL;
        }

        virtual bool get_next_feature(int32_t& index, float64_t& value, void* iterator)
        {
            SG_NOTIMPLEMENTED;
            return NULL;
        }

        virtual void free_feature_iterator(void* iterator)
        {
            SG_NOTIMPLEMENTED;
        }

        virtual inline int32_t get_nnz_features_for_vector(int32_t num)
        {
            SG_NOTIMPLEMENTED;
            return 0;
        }

        inline virtual EFeatureType get_feature_type()
        {
            return F_UNKNOWN;
        }

        inline virtual EFeatureClass get_feature_class()
        {
            return C_SPEC;
        }

        inline virtual int32_t get_num_vectors()
        {
            return num_strings;
        }

        inline virtual int32_t get_size()
        {
            return sizeof(float64_t);
        }

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

    protected:
        void obtain_kmer_spectrum(CStringFeatures<uint16_t>* str);

        void delete_kmer_spectrum();

    protected:
        bool use_normalization;
        int32_t num_strings;
        int32_t alphabet_size;

        int32_t spec_size;
        float64_t** k_spectrum;
};
}
#endif // _EXPLICITSPECFEATURES_H___