SHOGUN  v3.2.0
VowpalWabbit.h
浏览该文件的文档.
1 /*
2  * Copyright (c) 2009 Yahoo! Inc. All rights reserved. The copyrights
3  * embodied in the content of this file are licensed under the BSD
4  * (revised) open source license.
5  *
6  * This program is free software; you can redistribute it and/or modify
7  * it under the terms of the GNU General Public License as published by
8  * the Free Software Foundation; either version 3 of the License, or
9  * (at your option) any later version.
10  *
11  * Written (W) 2011 Shashwat Lal Das
12  * Adaptation of Vowpal Wabbit v5.1.
13  * Copyright (C) 2011 Berlin Institute of Technology and Max-Planck-Society.
14  */
15 
16 #ifndef _VOWPALWABBIT_H__
17 #define _VOWPALWABBIT_H__
18 
23 
26 
27 namespace shogun
28 {
39 {
40 public:
41 
44 
48  CVowpalWabbit();
49 
57 
62 
67 
72  void reinitialize_weights();
73 
82  void set_no_training(bool dont_train) { no_training = dont_train; }
83 
89  void set_adaptive(bool adaptive_learning);
90 
97  void set_exact_adaptive_norm(bool exact_adaptive);
98 
104  void set_num_passes(int32_t passes)
105  {
106  env->num_passes = passes;
107  }
108 
114  void load_regressor(char* file_name);
115 
122  void set_regressor_out(char* file_name, bool is_text = true);
123 
129  void set_prediction_out(char* file_name);
130 
137  void add_quadratic_pair(char* pair);
138 
144  virtual bool train_machine(CFeatures* feat = NULL);
145 
154 
164 
177  float32_t compute_exact_norm_quad(float32_t* weights, VwFeature& page_feature, v_array<VwFeature> &offer_features,
178  vw_size_t mask, float32_t g, float32_t& sum_abs_x);
179 
186  {
187  SG_REF(env);
188  return env;
189  }
190 
196  virtual const char* get_name() const { return "VowpalWabbit"; }
197 
202  virtual void set_learner();
203 
208 
209 private:
215  virtual void init(CStreamingVwFeatures* feat = NULL);
216 
224  virtual float32_t inline_l1_predict(VwExample* &ex);
225 
233  virtual float32_t inline_predict(VwExample* &ex);
234 
242  virtual float32_t finalize_prediction(float32_t ret);
243 
249  virtual void output_example(VwExample* &ex);
250 
256  virtual void print_update(VwExample* &ex);
257 
266  virtual void output_prediction(int32_t f, float32_t res, float32_t weight, v_array<char> tag);
267 
273  void set_verbose(bool verbose);
274 
275 protected:
278 
281 
284 
287 
288 private:
290  bool quiet;
291 
293  bool no_training;
294 
296  float32_t dump_interval;
298  float32_t sum_loss_since_last_dump;
300  float64_t old_weighted_examples;
301 
303  char* reg_name;
305  bool reg_dump_text;
306 
308  bool save_predictions;
310  int32_t prediction_fd;
311 };
312 
313 }
314 #endif // _VOWPALWABBIT_H__
uint32_t vw_size_t
vw_size_t typedef to work across platforms
Definition: vw_constants.h:24
CVwRegressor * reg
Regressor.
Definition: VowpalWabbit.h:286
Class OnlineLinearMachine is a generic interface for linear machines like classifiers which work thro...
void set_adaptive(bool adaptive_learning)
virtual CVwEnvironment * get_env()
Definition: VowpalWabbit.h:185
void set_prediction_out(char *file_name)
Class CVwEnvironment is the environment used by VW.
Definition: VwEnvironment.h:39
CVwLearner * get_learner()
Definition: VowpalWabbit.h:207
void set_num_passes(int32_t passes)
Definition: VowpalWabbit.h:104
Class v_array taken directly from JL's implementation.
CVwEnvironment * env
Environment for VW, i.e., globals.
Definition: VowpalWabbit.h:280
float32_t compute_exact_norm_quad(float32_t *weights, VwFeature &page_feature, v_array< VwFeature > &offer_features, vw_size_t mask, float32_t g, float32_t &sum_abs_x)
void load_regressor(char *file_name)
virtual void set_learner()
MACHINE_PROBLEM_TYPE(PT_BINARY)
double float64_t
Definition: common.h:48
float32_t compute_exact_norm(VwExample *&ex, float32_t &sum_abs_x)
Base class for all VW learners.
Definition: VwLearner.h:33
vw_size_t num_passes
Number of passes.
#define SG_REF(x)
Definition: SGRefObject.h:34
Regressor used by VW.
Definition: VwRegressor.h:35
void set_exact_adaptive_norm(bool exact_adaptive)
Example class for VW.
Definition: vw_example.h:56
virtual float32_t predict_and_finalize(VwExample *ex)
This class implements streaming features for use with VW.
void set_regressor_out(char *file_name, bool is_text=true)
float float32_t
Definition: common.h:47
One feature in VW.
Definition: vw_example.h:32
virtual const char * get_name() const
Definition: VowpalWabbit.h:196
void set_no_training(bool dont_train)
Definition: VowpalWabbit.h:82
virtual bool train_machine(CFeatures *feat=NULL)
all of classes and functions are contained in the shogun namespace
Definition: class_list.h:16
CStreamingVwFeatures * features
Features.
Definition: VowpalWabbit.h:277
The class Features is the base class of all feature objects.
Definition: Features.h:62
void add_quadratic_pair(char *pair)
Class CVowpalWabbit is the implementation of the online learning algorithm used in Vowpal Wabbit...
Definition: VowpalWabbit.h:38
CVwLearner * learner
Learner to use.
Definition: VowpalWabbit.h:283

SHOGUN Machine Learning Toolbox - Documentation