HypervolumeContribution.h
Go to the documentation of this file.
1/*!
2 *
3 *
4 * \brief Implements the frontend for the HypervolumeContribution algorithms, including the approximations
5 *
6 *
7 * \author O.Krause
8 * \date 2014-2016
9 *
10 *
11 * \par Copyright 1995-2017 Shark Development Team
12 *
13 * <BR><HR>
14 * This file is part of Shark.
15 * <https://shark-ml.github.io/Shark/>
16 *
17 * Shark is free software: you can redistribute it and/or modify
18 * it under the terms of the GNU Lesser General Public License as published
19 * by the Free Software Foundation, either version 3 of the License, or
20 * (at your option) any later version.
21 *
22 * Shark is distributed in the hope that it will be useful,
23 * but WITHOUT ANY WARRANTY; without even the implied warranty of
24 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
25 * GNU Lesser General Public License for more details.
26 *
27 * You should have received a copy of the GNU Lesser General Public License
28 * along with Shark. If not, see <http://www.gnu.org/licenses/>.
29 *
30 */
31#ifndef SHARK_ALGORITHMS_DIRECTSEARCH_HYPERVOLUMECONTRIBUTION_H
32#define SHARK_ALGORITHMS_DIRECTSEARCH_HYPERVOLUMECONTRIBUTION_H
33
34#include <shark/Algorithms/DirectSearch/Operators/Hypervolume/HypervolumeContribution2D.h>
35#include <shark/Algorithms/DirectSearch/Operators/Hypervolume/HypervolumeContribution3D.h>
36#include <shark/Algorithms/DirectSearch/Operators/Hypervolume/HypervolumeContributionMD.h>
37#include <shark/Algorithms/DirectSearch/Operators/Hypervolume/HypervolumeContributionApproximator.h>
38
39
40namespace shark {
41/// \brief Frontend for hypervolume contribution algorithms in m dimensions.
42///
43/// Depending on the dimensionality of the problem, one of the specialized algorithms is called.
44/// For large dimensionalities for which there are no specialized fast algorithms,
45/// the exponential time algorithm is called.
46/// Also a log-transformation of points is supported
47struct HypervolumeContribution {
48
49 /// \brief Default c'tor.
50 HypervolumeContribution() : m_useApproximation(false) {}
51
52 ///\brief True if the hypervolume approximation is to be used in dimensions > 3.
53 void useApproximation(bool useApproximation){
54 m_useApproximation = useApproximation;
55 }
56
57 double approximationEpsilon()const{
58 return m_approximationAlgorithm.epsilon();
59 }
60 double& approximationEpsilon(){
61 return m_approximationAlgorithm.epsilon();
62 }
63
64 double approximationDelta()const{
65 return m_approximationAlgorithm.delta();
66 }
67
68 double& approximationDelta(){
69 return m_approximationAlgorithm.delta();
70 }
71
72 template<typename Archive>
73 void serialize( Archive & archive, const unsigned int version ) {
74 archive & BOOST_SERIALIZATION_NVP(m_useApproximation);
75 archive & BOOST_SERIALIZATION_NVP(m_approximationAlgorithm);
76 }
77
78 /// \brief Returns the index of the points with smallest contribution as well as their contribution.
79 ///
80 /// \param [in] points The set \f$S\f$ of points from which to select the smallest contributor.
81 /// \param [in] k The number of points to select.
82 /// \param [in] ref The reference Point\f$\vec{r} \in \mathbb{R}^2\f$ for the hypervolume calculation, needs to fulfill: \f$ \forall s \in S: s \preceq \vec{r}\f$.
83 template<class Set, typename VectorType>
84 std::vector<KeyValuePair<double,std::size_t> > smallest(Set const& points, std::size_t k, VectorType const& ref)const{
85 SHARK_RUNTIME_CHECK(points.size() >= k, "There must be at least k points in the set");
86 SIZE_CHECK( points.begin()->size() == ref.size() );
87 std::size_t numObjectives = ref.size();
88 if(numObjectives == 2){
89 HypervolumeContribution2D algorithm;
90 return algorithm.smallest(points, k, ref);
91 }else if(numObjectives == 3){
92 HypervolumeContribution3D algorithm;
93 return algorithm.smallest(points, k, ref);
94 }else if(m_useApproximation){
95 return m_approximationAlgorithm.smallest(points, k, ref);
96 }else{
97 HypervolumeContributionMD algorithm;
98 return algorithm.smallest(points, k, ref);
99 }
100 }
101
102 /// \brief Returns the index of the points with largest contribution as well as their contribution.
103 ///
104 /// \param [in] points The set \f$S\f$ of points from which to select the largest contributor.
105 /// \param [in] k Number of points.
106 /// \param [in] ref The reference Point\f$\vec{r} \in \mathbb{R}^2\f$ for the hypervolume calculation, needs to fulfill: \f$ \forall s \in S: s \preceq \vec{r}\f$.
107 template<class Set, typename VectorType>
108 std::vector<KeyValuePair<double,std::size_t> > largest(Set const& points, std::size_t k, VectorType const& ref)const{
109 SHARK_RUNTIME_CHECK(points.size() >= k, "There must be at least k points in the set");
110 SIZE_CHECK( points.begin()->size() == ref.size() );
111 std::size_t numObjectives = ref.size();
112 if(numObjectives == 2){
113 HypervolumeContribution2D algorithm;
114 return algorithm.largest(points, k, ref);
115 }else if(numObjectives == 3){
116 HypervolumeContribution3D algorithm;
117 return algorithm.largest(points, k, ref);
118 }else{
119 SHARK_RUNTIME_CHECK(!m_useApproximation, "Largest not implemented for approximation algorithm");
120 HypervolumeContributionMD algorithm;
121 return algorithm.largest(points, k, ref);
122 }
123 }
124
125 /// \brief Returns the index of the points with smallest contribution as well as their contribution.
126 ///
127 /// As no reference point is given, the extremum points can not be computed and are never selected.
128 ///
129 /// \param [in] points The set \f$S\f$ of points from which to select the smallest contributor.
130 /// \param [in] k The number of points to select.
131 template<class Set>
132 std::vector<KeyValuePair<double,std::size_t> > smallest(Set const& points, std::size_t k)const{
133 SHARK_RUNTIME_CHECK(points.size() >= k, "There must be at least k points in the set");
134 std::size_t numObjectives = points[0].size();
135 if(numObjectives == 2){
136 HypervolumeContribution2D algorithm;
137 return algorithm.smallest(points, k);
138 }else if(numObjectives == 3){
139 HypervolumeContribution3D algorithm;
140 return algorithm.smallest(points, k);
141 }else if(m_useApproximation){
142 return m_approximationAlgorithm.smallest(points, k);
143 }else{
144 HypervolumeContributionMD algorithm;
145 return algorithm.smallest(points, k);
146 }
147 }
148
149 /// \brief Returns the index of the points with largest contribution as well as their contribution.
150 ///
151 /// As no reference point is given, the extremum points can not be computed and are never selected.
152 ///
153 /// \param [in] points The set \f$S\f$ of points from which to select the smallest contributor.
154 /// \param [in] k The number of points to select.
155 template<class Set>
156 std::vector<KeyValuePair<double,std::size_t> > largest(Set const& points, std::size_t k)const{
157 SHARK_RUNTIME_CHECK(points.size() >= k, "There must be at least k points in the set");
158 std::size_t numObjectives = points[0].size();
159 if(numObjectives == 2){
160 HypervolumeContribution2D algorithm;
161 return algorithm.largest(points, k);
162 }else if(numObjectives == 3){
163 HypervolumeContribution3D algorithm;
164 return algorithm.largest(points, k);
165 }else{
166 SHARK_RUNTIME_CHECK(!m_useApproximation, "Largest not implemented for approximation algorithm");
167 HypervolumeContributionMD algorithm;
168 return algorithm.largest(points, k);
169 }
170 }
171
172private:
173 bool m_useApproximation;
174 HypervolumeContributionApproximator m_approximationAlgorithm;
175};
176
177}
178#endif