SHOGUN  3.2.1
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Groups Pages
SoftMaxLikelihood.h
Go to the documentation of this file.
1 /*
2  * Copyright (c) The Shogun Machine Learning Toolbox
3  * Written (w) 2014 Parijat Mazumdar, Wu Lin
4  * All rights reserved.
5  *
6  * Redistribution and use in source and binary forms, with or without
7  * modification, are permitted provided that the following conditions are met:
8  *
9  * 1. Redistributions of source code must retain the above copyright notice, this
10  * list of conditions and the following disclaimer.
11  * 2. Redistributions in binary form must reproduce the above copyright notice,
12  * this list of conditions and the following disclaimer in the documentation
13  * and/or other materials provided with the distribution.
14  *
15  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
16  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
17  * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
18  * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
19  * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
20  * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
21  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
22  * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
23  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
24  * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
25  *
26  * The views and conclusions contained in the software and documentation are those
27  * of the authors and should not be interpreted as representing official policies,
28  * either expressed or implied, of the Shogun Development Team.
29  *
30  * Code adapted from
31  * https://gist.github.com/yorkerlin/8a36e8f9b298aa0246a4
32  * and
33  * GPstuff - Gaussian process models for Bayesian analysis
34  * http://becs.aalto.fi/en/research/bayes/gpstuff/
35  *
36  * The reference pseudo code is the algorithm 3.4 of the GPML textbook
37  *
38  */
39 
40 #ifndef _SOFTMAXLIKELIHOOD_H_
41 #define _SOFTMAXLIKELIHOOD_H_
42 
43 #include <shogun/lib/config.h>
44 
45 #ifdef HAVE_EIGEN3
46 
49 
50 namespace shogun
51 {
52 
55 {
59 };
60 
80 {
81 public:
84 
86  virtual ~CSoftMaxLikelihood();
87 
92  virtual const char* get_name() const { return "SoftMaxLikelihood"; }
93 
111  SGVector<float64_t> s2, const CLabels* lab=NULL) const;
112 
130  SGVector<float64_t> s2, const CLabels* lab=NULL) const;
131 
164  SGVector<float64_t> s2, const CLabels *lab=NULL) const;
165 
178  SGVector<float64_t> func) const;
179 
191  const CLabels* lab, SGVector<float64_t> func, index_t i) const;
192 
205  SGVector<float64_t> s2, const CLabels* lab) const
206  {
207  SG_ERROR("Not Implemented\n");
208  return SGVector<float64_t>();
209  }
210 
226  SGVector<float64_t> s2, const CLabels* lab, index_t i) const
227  {
228  SG_ERROR("Not Implemented\n");
229  return -1.0;
230  }
231 
247  SGVector<float64_t> s2, const CLabels* lab, index_t i) const
248  {
249  SG_ERROR("Not Implemented\n");
250  return -1.0;
251  }
252 
257  virtual bool supports_multiclass() const { return true; }
258 
264  virtual void set_num_samples(index_t num_samples);
265 
266 private:
268  void init();
270  index_t m_num_samples;
271 
285  SGVector<float64_t> predictive_helper(SGVector<float64_t> mu,
286  SGVector<float64_t> s2, const CLabels *lab, EMCSamplerType option) const;
287 
302  SGVector<float64_t> mc_sampler(index_t num_samples, SGVector<float64_t> mean,
304 
305 
314  SGVector<float64_t> get_log_probability_derivative1_f(const CLabels* lab, SGMatrix<float64_t> func) const;
315 
324  SGVector<float64_t> get_log_probability_derivative2_f(SGMatrix<float64_t> func) const;
325 
334  SGVector<float64_t> get_log_probability_derivative3_f(SGMatrix<float64_t> func) const;
335 };
336 }
337 #endif /* HAVE_EIGEN3 */
338 #endif /* _SOFTMAXLIKELIHOOD_H_ */

SHOGUN Machine Learning Toolbox - Documentation