SHOGUN  4.2.0
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules 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/config.h>
14 
15 #include <shogun/lib/common.h>
17 #include <shogun/base/Parameter.h>
18 #include <shogun/lib/DataType.h>
20 
21 #include <shogun/lib/SGString.h>
24 
25 namespace shogun
26 {
30 template <class T> class CStreamingStringFeatures : public CStreamingFeatures
31 {
32 public:
33 
42 
52  bool is_labelled,
53  int32_t size);
54 
60  virtual ~CStreamingStringFeatures();
61 
71  virtual void set_vector_reader();
72 
82  virtual void set_vector_and_label_reader();
83 
90  void use_alphabet(EAlphabet alpha);
91 
98  void use_alphabet(CAlphabet* alpha);
99 
107  void set_remap(CAlphabet* ascii_alphabet, CAlphabet* binary_alphabet);
108 
116  void set_remap(EAlphabet ascii_alphabet=DNA, EAlphabet binary_alphabet=RAWDNA);
117 
123 
131 
137  virtual void start_parser();
138 
144  virtual void end_parser();
145 
154  virtual bool get_next_example();
155 
162 
170  virtual float64_t get_label();
171 
178  virtual void release_example();
179 
185  virtual int32_t get_vector_length();
186 
192  virtual EFeatureType get_feature_type() const;
193 
199  virtual EFeatureClass get_feature_class() const;
200 
206  virtual CFeatures* duplicate() const;
207 
213  virtual const char* get_name() const { return "StreamingStringFeatures"; }
214 
220  virtual int32_t get_num_vectors() const;
221 
227  virtual int32_t get_num_features();
228 
229 private:
230 
235  void init();
236 
244  void init(CStreamingFile *file, bool is_labelled, int32_t size);
245 
246 protected:
247 
249  CInputParser<T> parser;
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__
CAlphabet * alphabet
Alphabet to use.
float64_t current_label
The label of the current example, if applicable.
virtual CFeatures * duplicate() const
DNA - letters A,C,G,T.
Definition: Alphabet.h:26
int32_t num_symbols
Number of symbols.
virtual const char * get_name() const
RAWDNA - letters 0,1,2,3.
Definition: Alphabet.h:29
EAlphabet
Alphabet of charfeatures/observations.
Definition: Alphabet.h:23
virtual EFeatureType get_feature_type() const
The class Alphabet implements an alphabet and alphabet utility functions.
Definition: Alphabet.h:91
CInputParser< T > parser
The parser object, which reads from input and returns parsed example objects.
EFeatureClass
shogun feature class
Definition: FeatureTypes.h:38
A Streaming File access class.
Definition: StreamingFile.h:34
virtual EFeatureClass get_feature_class() const
shogun string
bool remap_to_bin
Whether remapping must be done.
This class implements streaming features as strings.
double float64_t
Definition: common.h:50
long double floatmax_t
Definition: common.h:51
T * current_string
The current example's string as a T*.
SGString< T > current_sgstring
The current example's string as an SGString
EFeatureType
shogun feature type
Definition: FeatureTypes.h:19
all of classes and functions are contained in the shogun namespace
Definition: class_list.h:18
The class Features is the base class of all feature objects.
Definition: Features.h:68
Streaming features are features which are used for online algorithms.
CAlphabet * alpha_bin
If remapping is enabled, this is the target alphabet.
bool has_labels
Whether examples are labelled or not.
void set_remap(CAlphabet *ascii_alphabet, CAlphabet *binary_alphabet)
CAlphabet * alpha_ascii
If remapping is enabled, this is the source alphabet.
int32_t current_length
The length of the current string.

SHOGUN Machine Learning Toolbox - Documentation