HashedWDFeatures.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) 2010 Soeren Sonnenburg
 * Copyright (C) 2010 Berlin Institute of Technology
 */

#ifndef _HASHEDWDFEATURES_H___
#define _HASHEDWDFEATURES_H___

#include <shogun/lib/common.h>
#include <shogun/features/DotFeatures.h>
#include <shogun/features/StringFeatures.h>
#include <shogun/lib/Hash.h>

namespace shogun
{
template<class ST> class CStringFeatures;

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

    CHashedWDFeatures(CStringFeatures<uint8_t>* str, int32_t start_order,
            int32_t order, int32_t from_order, int32_t hash_bits=12);

    CHashedWDFeatures(const CHashedWDFeatures & orig);

    virtual ~CHashedWDFeatures();

    inline virtual int32_t get_dim_feature_space() const
    {
        return w_dim;
    }

    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);

    virtual inline int32_t get_nnz_features_for_vector(int32_t num)
    {
        int32_t vlen=-1;
        bool free_vec;
        uint8_t* vec=strings->get_feature_vector(num, vlen, free_vec);
        strings->free_feature_vector(vec, num, free_vec);
        return degree*vlen;
    }

    #ifndef DOXYGEN_SHOULD_SKIP_THIS

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

        int32_t index;

    };
    #endif

    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 CFeatures* duplicate() const;

    inline virtual EFeatureType get_feature_type()
    {
        return F_UNKNOWN;
    }

    inline virtual EFeatureClass get_feature_class()
    {
        return C_WD;
    }

    inline virtual int32_t get_num_vectors() const
    {
        return num_strings;
    }

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

    void set_normalization_const(float64_t n=0);

    inline float64_t get_normalization_const()
    {
        return normalization_const;
    }

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

protected:

    void set_wd_weights();

protected:
    CStringFeatures<uint8_t>* strings;

    int32_t degree;
    int32_t start_degree;
    int32_t from_degree;
    int32_t string_length;
    int32_t num_strings;
    int32_t alphabet_size;
    int32_t w_dim;
    int32_t partial_w_dim;
    float64_t* wd_weights;
    uint32_t mask;
    int32_t m_hash_bits;

    float64_t normalization_const;
};
}
#endif // _HASHEDWDFEATURES_H___