SHOGUN  4.2.0
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules Pages
EMMixtureModel.cpp
Go to the documentation of this file.
1 /*
2  * Copyright (c) The Shogun Machine Learning Toolbox
3  * Written (w) 2014 Parijat Mazumdar
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 
34 
35 using namespace shogun;
36 
39 { }
40 
42 { }
43 
45 {
46  float64_t log_likelihood=0;
47  // for each data point
48  for (int32_t i=0;i<data.alpha.num_rows;i++)
49  {
51  // for each component
52  for (int32_t j=0;j<data.alpha.num_cols;j++)
53  {
55  alpha_ij[j]=CMath::log(data.weights[j])+jth_component->get_log_likelihood_example(i);
56  SG_UNREF(jth_component);
57  };
58 
59  float64_t normalize=CMath::log_sum_exp(alpha_ij);
60  log_likelihood+=normalize;
61 
62  // fill row of alpha
63  for (int32_t j=0;j<data.alpha.num_cols;j++)
64  data.alpha(i,j)=CMath::exp(alpha_ij[j]-normalize);
65  }
66 
67  return log_likelihood;
68 }
69 
71 {
72  // for each component
73  float64_t* alpha_j=NULL;
74  float64_t sum_weights=0;
75  for (int32_t j=0;j<data.alpha.num_cols;j++)
76  {
78 
79  // update mean covariance of components
80  alpha_j=data.alpha.matrix+j*data.alpha.num_rows;
81  float64_t weight_j=jth_component->update_params_em(alpha_j,data.alpha.num_rows);
82 
83  // update weights
84  sum_weights+=weight_j;
85  data.weights[j]=weight_j;
86 
87  SG_UNREF(jth_component);
88  }
89 
90  // update weights - normalization
91  for (int32_t j=0;j<data.alpha.num_cols;j++)
92  data.weights[j]/=sum_weights;
93 }
This is the base class for Expectation Maximization (EM). EM for various purposes can be derived from...
Definition: EMBase.h:44
SGVector< float64_t > weights
Definition: MixModelData.h:52
static CDistribution * obtain_from_generic(CSGObject *object)
SGMatrix< float64_t > alpha
Definition: MixModelData.h:48
index_t num_cols
Definition: SGMatrix.h:376
CDynamicObjectArray * components
Definition: MixModelData.h:50
Base class Distribution from which all methods implementing a distribution are derived.
Definition: Distribution.h:44
index_t num_rows
Definition: SGMatrix.h:374
double float64_t
Definition: common.h:50
virtual float64_t update_params_em(float64_t *alpha_k, int32_t len)
static T log_sum_exp(SGVector< T > values)
Definition: Math.h:1242
#define SG_UNREF(x)
Definition: SGObject.h:55
all of classes and functions are contained in the shogun namespace
Definition: class_list.h:18
This structure is used for storing data required for using the generic Expectation Maximization (EM) ...
Definition: MixModelData.h:45
static float64_t exp(float64_t x)
Definition: Math.h:621
static float64_t log(float64_t v)
Definition: Math.h:922
CSGObject * get_element(int32_t index) const
virtual float64_t expectation_step()
virtual float64_t get_log_likelihood_example(int32_t num_example)=0
virtual void maximization_step()

SHOGUN Machine Learning Toolbox - Documentation