SHOGUN  6.0.0
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Friends Macros Modules
Perceptron.cpp
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) 1999-2009 Soeren Sonnenburg
8  * Copyright (C) 1999-2009 Fraunhofer Institute FIRST and Max-Planck-Society
9  */
10 
12 #include <shogun/labels/Labels.h>
15 #include <shogun/lib/Signal.h>
16 
17 using namespace shogun;
18 
20 : CLinearMachine(), learn_rate(0.1), max_iter(1000), m_initialize_hyperplane(true)
21 {
22 }
23 
25 : CLinearMachine(), learn_rate(0.1), max_iter(1000), m_initialize_hyperplane(true)
26 {
27  set_features(traindat);
28  set_labels(trainlab);
29 }
30 
32 {
33 }
34 
36 {
39 
40  if (data)
41  {
42  if (!data->has_property(FP_DOT))
43  SG_ERROR("Specified features are not of type CDotFeatures\n")
44  set_features((CDotFeatures*) data);
45  }
46 
48  bool converged=false;
49  int32_t iter=0;
50  SGVector<int32_t> train_labels=((CBinaryLabels*) m_labels)->get_int_labels();
51  int32_t num_feat=features->get_dim_feature_space();
52  int32_t num_vec=features->get_num_vectors();
53 
54  ASSERT(num_vec==train_labels.vlen)
55  float64_t* output=SG_MALLOC(float64_t, num_vec);
56 
58  if (m_initialize_hyperplane)
59  {
60  w = SGVector<float64_t>(num_feat);
61  //start with uniform w, bias=0
62  bias=0;
63  for (int32_t i=0; i<num_feat; i++)
64  w.vector[i]=1.0/num_feat;
65  }
66 
68 
69  //loop till we either get everything classified right or reach max_iter
70  while (!(CSignal::cancel_computations()) && (!converged && iter<max_iter))
71  {
72  converged=true;
73  for (int32_t i=0; i<num_vec; i++)
74  {
75  output[i] = features->dense_dot(i, w.vector, w.vlen) + bias;
76 
77  if (CMath::sign<float64_t>(output[i]) != train_labels.vector[i])
78  {
79  converged=false;
80  bias+=learn_rate*train_labels.vector[i];
81  features->add_to_dense_vec(learn_rate*train_labels.vector[i], i, w.vector, w.vlen);
82  }
83  }
84 
85  iter++;
86  }
87 
88  if (converged)
89  SG_INFO("Perceptron algorithm converged after %d iterations.\n", iter)
90  else
91  SG_WARNING("Perceptron algorithm did not converge after %d iterations.\n", max_iter)
92 
93  SG_FREE(output);
94 
95  set_w(w);
96 
97  return converged;
98 }
99 
100 void CPerceptron::set_initialize_hyperplane(bool initialize_hyperplane)
101 {
102  m_initialize_hyperplane = initialize_hyperplane;
103 }
104 
106 {
107  return m_initialize_hyperplane;
108 }
#define SG_INFO(...)
Definition: SGIO.h:117
virtual ELabelType get_label_type() const =0
binary labels +1/-1
Definition: LabelTypes.h:18
virtual void set_w(const SGVector< float64_t > src_w)
The class Labels models labels, i.e. class assignments of objects.
Definition: Labels.h:43
virtual float64_t dense_dot(int32_t vec_idx1, const float64_t *vec2, int32_t vec2_len)=0
bool get_initialize_hyperplane()
get if the hyperplane should be initialized
Definition: Perceptron.cpp:105
virtual int32_t get_num_vectors() const =0
CLabels * m_labels
Definition: Machine.h:365
#define SG_ERROR(...)
Definition: SGIO.h:128
virtual void add_to_dense_vec(float64_t alpha, int32_t vec_idx1, float64_t *vec2, int32_t vec2_len, bool abs_val=false)=0
Features that support dot products among other operations.
Definition: DotFeatures.h:44
virtual int32_t get_dim_feature_space() const =0
index_t vlen
Definition: SGVector.h:545
#define ASSERT(x)
Definition: SGIO.h:200
static void clear_cancel()
Definition: Signal.cpp:126
double float64_t
Definition: common.h:60
void set_initialize_hyperplane(bool initialize_hyperplane)
set if the hyperplane should be initialized
Definition: Perceptron.cpp:100
virtual void set_features(CDotFeatures *feat)
Class LinearMachine is a generic interface for all kinds of linear machines like classifiers.
Definition: LinearMachine.h:63
virtual SGVector< float64_t > get_w() const
static bool cancel_computations()
Definition: Signal.h:111
CDotFeatures * features
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
float64_t learn_rate
Definition: Perceptron.h:89
virtual bool train_machine(CFeatures *data=NULL)
Definition: Perceptron.cpp:35
Binary Labels for binary classification.
Definition: BinaryLabels.h:37
#define SG_WARNING(...)
Definition: SGIO.h:127
bool has_property(EFeatureProperty p) const
Definition: Features.cpp:295
virtual void set_labels(CLabels *lab)
Definition: Machine.cpp:65
virtual ~CPerceptron()
Definition: Perceptron.cpp:31

SHOGUN Machine Learning Toolbox - Documentation