ImplicitWeightedSpecFeatures.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) 2009-2010 Soeren Sonnenburg
 * Copyright (C) 2009 Fraunhofer Institute FIRST and Max-Planck-Society
 * Copyright (C) 2010 Berlin Institute of Technology
 */

#ifndef _IMPLICITSPECFEATURES_H___
#define _IMPLICITSPECFEATURES_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 CImplicitWeightedSpecFeatures : public CDotFeatures
{
    public:

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

        CImplicitWeightedSpecFeatures(const CImplicitWeightedSpecFeatures & orig);

        virtual ~CImplicitWeightedSpecFeatures();

        virtual CFeatures* duplicate() const;

        inline virtual int32_t get_dim_feature_space()
        {
            return spec_size;
        }

        virtual float64_t dot(int32_t vec_idx1, int32_t vec_idx2);

        virtual float64_t dense_dot(int32_t vec_idx1, 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);

        virtual inline int32_t get_nnz_features_for_vector(int32_t num)
        {
            int32_t vlen=-1;
            bool free_vec;
            uint16_t* vec1=strings->get_feature_vector(num, vlen, free_vec);
            strings->free_feature_vector(vec1, num, free_vec);
            int32_t nnz=0;
            for (int32_t i=1; i<=degree; i++)
                nnz+=CMath::min(CMath::pow(alphabet_size,i), vlen);
            return nnz;
        }

        inline virtual EFeatureType get_feature_type()
        {
            return F_UNKNOWN;
        }

        inline virtual EFeatureClass get_feature_class()
        {
            return C_WEIGHTEDSPEC;
        }

        inline virtual int32_t get_num_vectors()
        {
            return num_strings;
        }

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

        bool set_wd_weights();

        bool set_weights(float64_t* w, int32_t d);

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

            int32_t offs;
            int32_t d;
            int32_t j;
            uint8_t mask;
            float64_t alpha;
        };

        virtual void* get_feature_iterator(int32_t vector_index);

        virtual bool get_next_feature(int32_t& index, float64_t& value, void* iterator);

        virtual void free_feature_iterator(void* iterator);

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

    protected:
        void compute_normalization_const();

    protected:
        CStringFeatures<uint16_t>* strings;

        float64_t* normalization_factors;
        int32_t num_strings;
        int32_t alphabet_size;

        int32_t degree;
        int32_t spec_size;

        float64_t* spec_weights;
};
}
#endif // _IMPLICITSPECFEATURES_H___