Point Cloud Library (PCL)  1.12.0
fern_trainer.h
1 /*
2  * Software License Agreement (BSD License)
3  *
4  * Point Cloud Library (PCL) - www.pointclouds.org
5  * Copyright (c) 2010-2011, Willow Garage, Inc.
6  *
7  * All rights reserved.
8  *
9  * Redistribution and use in source and binary forms, with or without
10  * modification, are permitted provided that the following conditions
11  * are met:
12  *
13  * * Redistributions of source code must retain the above copyright
14  * notice, this list of conditions and the following disclaimer.
15  * * Redistributions in binary form must reproduce the above
16  * copyright notice, this list of conditions and the following
17  * disclaimer in the documentation and/or other materials provided
18  * with the distribution.
19  * * Neither the name of Willow Garage, Inc. nor the names of its
20  * contributors may be used to endorse or promote products derived
21  * from this software without specific prior written permission.
22  *
23  * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
24  * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
25  * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
26  * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
27  * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
28  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
29  * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
30  * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
31  * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
33  * ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
34  * POSSIBILITY OF SUCH DAMAGE.
35  *
36  */
37 
38 #pragma once
39 
40 #include <pcl/common/common.h>
41 #include <pcl/ml/feature_handler.h>
42 #include <pcl/ml/ferns/fern.h>
43 #include <pcl/ml/stats_estimator.h>
44 
45 #include <vector>
46 
47 namespace pcl {
48 
49 /** Trainer for a Fern. */
50 template <class FeatureType,
51  class DataSet,
52  class LabelType,
53  class ExampleIndex,
54  class NodeType>
56 
57 public:
58  /** Constructor. */
59  FernTrainer();
60 
61  /** Destructor. */
62  virtual ~FernTrainer();
63 
64  /** Sets the feature handler used to create and evaluate features.
65  *
66  * \param[in] feature_handler the feature handler
67  */
68  inline void
71  {
72  feature_handler_ = &feature_handler;
73  }
74 
75  /** Sets the object for estimating the statistics for tree nodes.
76  *
77  * \param[in] stats_estimator the statistics estimator
78  */
79  inline void
82  {
83  stats_estimator_ = &stats_estimator;
84  }
85 
86  /** Sets the maximum depth of the learned tree.
87  *
88  * \param[in] fern_depth maximum depth of the learned tree
89  */
90  inline void
91  setFernDepth(const std::size_t fern_depth)
92  {
93  fern_depth_ = fern_depth;
94  }
95 
96  /** Sets the number of features used to find optimal decision features.
97  *
98  * \param[in] num_of_features the number of features
99  */
100  inline void
101  setNumOfFeatures(const std::size_t num_of_features)
102  {
103  num_of_features_ = num_of_features;
104  }
105 
106  /** Sets the number of thresholds tested for finding the optimal decision
107  * threshold on the feature responses.
108  *
109  * \param[in] num_of_threshold the number of thresholds
110  */
111  inline void
112  setNumOfThresholds(const std::size_t num_of_threshold)
113  {
114  num_of_thresholds_ = num_of_threshold;
115  }
116 
117  /** Sets the input data set used for training.
118  *
119  * \param[in] data_set the data set used for training
120  */
121  inline void
122  setTrainingDataSet(DataSet& data_set)
123  {
124  data_set_ = data_set;
125  }
126 
127  /** Example indices that specify the data used for training.
128  *
129  * \param[in] examples the examples
130  */
131  inline void
132  setExamples(std::vector<ExampleIndex>& examples)
133  {
134  examples_ = examples;
135  }
136 
137  /** Sets the label data corresponding to the example data.
138  *
139  * \param[in] label_data the label data
140  */
141  inline void
142  setLabelData(std::vector<LabelType>& label_data)
143  {
144  label_data_ = label_data;
145  }
146 
147  /** Trains a decision tree using the set training data and settings.
148  *
149  * \param[out] fern destination for the trained tree
150  */
151  void
152  train(Fern<FeatureType, NodeType>& fern);
153 
154 protected:
155  /** Creates uniformely distrebuted thresholds over the range of the supplied
156  * values.
157  *
158  * \param[in] num_of_thresholds the number of thresholds to create
159  * \param[in] values the values for estimating the expected value range
160  * \param[out] thresholds the resulting thresholds
161  */
162  static void
163  createThresholdsUniform(const std::size_t num_of_thresholds,
164  std::vector<float>& values,
165  std::vector<float>& thresholds);
166 
167 private:
168  /** Desired depth of the learned fern. */
169  std::size_t fern_depth_;
170  /** Number of features used to find optimal decision features. */
171  std::size_t num_of_features_;
172  /** Number of thresholds. */
173  std::size_t num_of_thresholds_;
174 
175  /** FeatureHandler instance, responsible for creating and evaluating features. */
177  /** StatsEstimator instance, responsible for gathering stats about a node. */
179 
180  /** The training data set. */
181  DataSet data_set_;
182  /** The label data. */
183  std::vector<LabelType> label_data_;
184  /** The example data. */
185  std::vector<ExampleIndex> examples_;
186 };
187 
188 } // namespace pcl
189 
190 #include <pcl/ml/impl/ferns/fern_trainer.hpp>
Utility class interface which is used for creating and evaluating features.
Class representing a Fern.
Definition: fern.h:49
Trainer for a Fern.
Definition: fern_trainer.h:55
void setTrainingDataSet(DataSet &data_set)
Sets the input data set used for training.
Definition: fern_trainer.h:122
void setFernDepth(const std::size_t fern_depth)
Sets the maximum depth of the learned tree.
Definition: fern_trainer.h:91
void setNumOfFeatures(const std::size_t num_of_features)
Sets the number of features used to find optimal decision features.
Definition: fern_trainer.h:101
void setStatsEstimator(pcl::StatsEstimator< LabelType, NodeType, DataSet, ExampleIndex > &stats_estimator)
Sets the object for estimating the statistics for tree nodes.
Definition: fern_trainer.h:80
void setNumOfThresholds(const std::size_t num_of_threshold)
Sets the number of thresholds tested for finding the optimal decision threshold on the feature respon...
Definition: fern_trainer.h:112
void setFeatureHandler(pcl::FeatureHandler< FeatureType, DataSet, ExampleIndex > &feature_handler)
Sets the feature handler used to create and evaluate features.
Definition: fern_trainer.h:69
void setLabelData(std::vector< LabelType > &label_data)
Sets the label data corresponding to the example data.
Definition: fern_trainer.h:142
void setExamples(std::vector< ExampleIndex > &examples)
Example indices that specify the data used for training.
Definition: fern_trainer.h:132
Define standard C methods and C++ classes that are common to all methods.
#define PCL_EXPORTS
Definition: pcl_macros.h:323