|
SHOGUN v0.9.3
|
00001 /* 00002 * This program is free software; you can redistribute it and/or modify 00003 * it under the terms of the GNU General Public License as published by 00004 * the Free Software Foundation; either version 3 of the License, or 00005 * (at your option) any later version. 00006 * 00007 * Written (W) 2009-2010 Soeren Sonnenburg 00008 * Copyright (C) 2009 Fraunhofer Institute FIRST and Max-Planck-Society 00009 * Copyright (C) 2010 Berlin Institute of Technology 00010 */ 00011 00012 #ifndef _DOTFEATURES_H___ 00013 #define _DOTFEATURES_H___ 00014 00015 #include "lib/common.h" 00016 #include "lib/Time.h" 00017 #include "lib/Mathematics.h" 00018 #include "features/Features.h" 00019 00020 namespace shogun 00021 { 00043 class CDotFeatures : public CFeatures 00044 { 00045 public: 00046 00051 CDotFeatures(int32_t size=0) : CFeatures(size), combined_weight(1.0) 00052 { 00053 set_property(FP_DOT); 00054 } 00055 00057 CDotFeatures(const CDotFeatures & orig) : 00058 CFeatures(orig), combined_weight(orig.combined_weight) {} 00059 00064 CDotFeatures(CFile* loader) : CFeatures(loader) {} 00065 00066 virtual ~CDotFeatures() { } 00067 00075 virtual int32_t get_dim_feature_space()=0; 00076 00083 virtual float64_t dot(int32_t vec_idx1, int32_t vec_idx2)=0; 00084 00091 virtual float64_t dense_dot(int32_t vec_idx1, float64_t* vec2, int32_t vec2_len)=0; 00092 00101 virtual void add_to_dense_vec(float64_t alpha, int32_t vec_idx1, float64_t* vec2, int32_t vec2_len, bool abs_val=false)=0; 00102 00116 virtual void dense_dot_range(float64_t* output, int32_t start, int32_t stop, float64_t* alphas, float64_t* vec, int32_t dim, float64_t b); 00117 00129 virtual void dense_dot_range_subset(int32_t* sub_index, int32_t num, float64_t* output, float64_t* alphas, float64_t* vec, int32_t dim, float64_t b); 00130 00133 static void* dense_dot_range_helper(void* p); 00134 00142 virtual int32_t get_nnz_features_for_vector(int32_t num)=0; 00143 00148 inline float64_t get_combined_feature_weight() { return combined_weight; } 00149 00154 inline void set_combined_feature_weight(float64_t nw) { combined_weight=nw; } 00155 00163 virtual void get_feature_matrix(float64_t** dst, int32_t* num_feat, int32_t* num_vec); 00164 00172 virtual void get_feature_vector(float64_t** dst, int32_t* len, int32_t num); 00173 00175 void benchmark_add_to_dense_vector(int32_t repeats=5); 00176 00178 void benchmark_dense_dot_range(int32_t repeats=5); 00179 00189 virtual void* get_feature_iterator(int32_t vector_index)=0; 00190 00201 virtual bool get_next_feature(int32_t& index, float64_t& value, void* iterator)=0; 00202 00208 virtual void free_feature_iterator(void* iterator)=0; 00209 00210 protected: 00217 inline void display_progress(int32_t start, int32_t stop, int32_t v) 00218 { 00219 int32_t num_vectors=stop-start; 00220 int32_t i=v-start; 00221 00222 if ( (i% (num_vectors/100+1))== 0) 00223 SG_PROGRESS(v, 0.0, num_vectors-1); 00224 } 00225 00226 #ifdef HAVE_BOOST_SERIALIZATION 00227 friend class ::boost::serialization::access; 00228 // When the class Archive corresponds to an output archive, the 00229 // & operator is defined similar to <<. Likewise, when the class Archive 00230 // is a type of input archive the & operator is defined similar to >>. 00231 template<class Archive> 00232 void serialize(Archive & ar, const unsigned int archive_version) 00233 { 00234 00235 SG_DEBUG("archiving DotFeatures\n"); 00236 00237 ar & ::boost::serialization::base_object<CFeatures>(*this); 00238 ar & combined_weight; 00239 00240 SG_DEBUG("done with DotFeatures\n"); 00241 } 00242 00243 #endif //HAVE_BOOST_SERIALIZATION 00244 00245 protected: 00246 00248 float64_t combined_weight; 00249 }; 00250 } 00251 #endif // _DOTFEATURES_H___