StreamingHashedSparseFeatures.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) 2013 Evangelos Anagnostopoulos
 * Copyright (C) 2013 Evangelos Anagnostopoulos
 */

#ifndef _STREAMING_HASHED_SPARSEFEATURES__H__
#define _STREAMING_HASHED_SPARSEFEATURES__H__

#include <shogun/lib/config.h>

#include <shogun/features/SparseFeatures.h>
#include <shogun/features/streaming/StreamingDotFeatures.h>
#include <shogun/io/streaming/InputParser.h>

namespace shogun
{
class CStreamingDotFeatures;

template <class ST> class CStreamingHashedSparseFeatures : public CStreamingDotFeatures
{
public:
    CStreamingHashedSparseFeatures();

    CStreamingHashedSparseFeatures (CStreamingFile* file, bool is_labelled, int32_t size,
                int32_t d = 512, bool use_quadr = false, bool keep_lin_terms = true);

    CStreamingHashedSparseFeatures (CSparseFeatures<ST>* dot_features, int32_t d = 512,
                bool use_quadr = false, bool keep_lin_terms = true, float64_t* lab = NULL);

    virtual ~CStreamingHashedSparseFeatures ();

    virtual float32_t dot(CStreamingDotFeatures* df);

    virtual float32_t dense_dot(const float32_t* vec2, int32_t vec2_len);

    virtual void add_to_dense_vec(float32_t alpha, float32_t* vec2,
            int32_t vec2_len, bool abs_val = false);

    virtual int32_t get_dim_feature_space() const;

    virtual const char* get_name() const;

    virtual int32_t get_num_vectors() const;

    virtual CFeatures* duplicate() const;

    virtual void set_vector_reader();

    virtual void set_vector_and_label_reader();

    virtual EFeatureType get_feature_type() const;

    virtual EFeatureClass get_feature_class() const;

    virtual void start_parser();

    virtual void end_parser();

    virtual float64_t get_label();

    virtual bool get_next_example();

    virtual void release_example();

    virtual int32_t get_num_features();

    SGSparseVector<ST> get_vector();

private:
    void init(CStreamingFile* file, bool is_labelled, int32_t size,
        int32_t d, bool use_quadr, bool keep_lin_terms);

protected:

    int32_t dim;

    SGSparseVector<ST> current_vector;

    CInputParser<SGSparseVectorEntry<ST> > parser;

    float64_t current_label;

    bool use_quadratic;

    bool keep_linear_terms;
};
}

#endif // _STREAMING_HASHED_SPARSEFEATURES__H__