SHOGUN  v3.0.0
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
StreamingStringFeatures.h
Go to the documentation of this file.
1 /*
2  * This program is free software; you can redistribute it and/or modify
3  * it under the terms of the GNU General Public License as published by
4  * the Free Software Foundation; either version 3 of the License, or
5  * (at your option) any later version.
6  *
7  * Written (W) 2011 Shashwat Lal Das
8  * Copyright (C) 2011 Berlin Institute of Technology and Max-Planck-Society
9  */
10 #ifndef _STREAMING_STRINGFEATURES__H__
11 #define _STREAMING_STRINGFEATURES__H__
12 
13 #include <shogun/lib/common.h>
15 #include <shogun/base/Parameter.h>
16 #include <shogun/lib/DataType.h>
18 
21 
22 namespace shogun
23 {
27 template <class T> class CStreamingStringFeatures : public CStreamingFeatures
28 {
29 public:
30 
39 
49  bool is_labelled,
50  int32_t size);
51 
57  virtual ~CStreamingStringFeatures();
58 
68  virtual void set_vector_reader();
69 
79  virtual void set_vector_and_label_reader();
80 
87  void use_alphabet(EAlphabet alpha);
88 
95  void use_alphabet(CAlphabet* alpha);
96 
104  void set_remap(CAlphabet* ascii_alphabet, CAlphabet* binary_alphabet);
105 
113  void set_remap(EAlphabet ascii_alphabet=DNA, EAlphabet binary_alphabet=RAWDNA);
114 
120 
128 
134  virtual void start_parser();
135 
141  virtual void end_parser();
142 
151  virtual bool get_next_example();
152 
159 
167  virtual float64_t get_label();
168 
175  virtual void release_example();
176 
182  virtual int32_t get_vector_length();
183 
189  virtual EFeatureType get_feature_type() const;
190 
196  virtual EFeatureClass get_feature_class() const;
197 
203  virtual CFeatures* duplicate() const;
204 
210  virtual const char* get_name() const { return "StreamingStringFeatures"; }
211 
217  virtual int32_t get_num_vectors() const;
218 
224  virtual int32_t get_num_features();
225 
226 private:
227 
232  void init();
233 
241  void init(CStreamingFile *file, bool is_labelled, int32_t size);
242 
243 protected:
244 
246  CInputParser<T> parser;
247 
250 
253 
256 
259 
262 
265 
267  int32_t current_length;
268 
271 
274 
277 
279  int32_t num_symbols;
280 };
281 
282 }
283 #endif // _STREAMING_STRINGFEATURES__H__

SHOGUN Machine Learning Toolbox - Documentation