Shark machine learning library
Installation
Tutorials
Benchmarks
Documentation
Quick references
Class list
Global functions
include
shark
ObjectiveFunctions
Benchmarks
ZDT6.h
Go to the documentation of this file.
1
//===========================================================================
2
/*!
3
*
4
*
5
* \brief Multi-objective optimization benchmark function ZDT6
6
*
7
* The function is described in
8
*
9
* Eckart Zitzler, Kalyanmoy Deb, and Lothar Thiele. Comparison of
10
* Multiobjective Evolutionary Algorithms: Empirical
11
* Results. Evolutionary Computation 8(2):173-195, 2000
12
*
13
*
14
*
15
* \author -
16
* \date -
17
*
18
*
19
* \par Copyright 1995-2017 Shark Development Team
20
*
21
* <BR><HR>
22
* This file is part of Shark.
23
* <https://shark-ml.github.io/Shark/>
24
*
25
* Shark is free software: you can redistribute it and/or modify
26
* it under the terms of the GNU Lesser General Public License as published
27
* by the Free Software Foundation, either version 3 of the License, or
28
* (at your option) any later version.
29
*
30
* Shark is distributed in the hope that it will be useful,
31
* but WITHOUT ANY WARRANTY; without even the implied warranty of
32
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
33
* GNU Lesser General Public License for more details.
34
*
35
* You should have received a copy of the GNU Lesser General Public License
36
* along with Shark. If not, see <http://www.gnu.org/licenses/>.
37
*
38
*/
39
//===========================================================================
40
41
#ifndef SHARK_OBJECTIVEFUNCTIONS_BENCHMARK_ZDT6_H
42
#define SHARK_OBJECTIVEFUNCTIONS_BENCHMARK_ZDT6_H
43
44
#include <
shark/ObjectiveFunctions/AbstractObjectiveFunction.h
>
45
#include <
shark/ObjectiveFunctions/BoxConstraintHandler.h
>
46
47
namespace
shark
{
namespace
benchmarks{
48
/*! \brief Multi-objective optimization benchmark function ZDT6
49
*
50
* The function is described in
51
*
52
* Eckart Zitzler, Kalyanmoy Deb, and Lothar Thiele. Comparison of
53
* Multiobjective Evolutionary Algorithms: Empirical
54
* Results. Evolutionary Computation 8(2):173-195, 2000
55
* \ingroup benchmarks
56
*/
57
struct
ZDT6
:
public
MultiObjectiveFunction
58
{
59
60
ZDT6
(std::size_t numVariables = 0) : m_handler(numVariables,0,1){
61
announceConstraintHandler
(&m_handler);
62
}
63
64
/// \brief From INameable: return the class name.
65
std::string
name
()
const
66
{
return
"ZDT6"
; }
67
68
std::size_t
numberOfObjectives
()
const
{
69
return
2;
70
}
71
72
std::size_t
numberOfVariables
()
const
{
73
return
m_handler.
dimensions
();
74
}
75
76
bool
hasScalableDimensionality
()
const
{
77
return
true
;
78
}
79
80
/// \brief Adjusts the number of variables if the function is scalable.
81
/// \param [in] numberOfVariables The new dimension.
82
void
setNumberOfVariables
( std::size_t
numberOfVariables
){
83
m_handler.
setBounds
(
numberOfVariables
,0,1);
84
}
85
86
// std::vector<double> evaluate( const point_type & x ) {
87
ResultType
eval
(
const
SearchPointType
& x )
const
{
88
m_evaluationCounter
++;
89
90
ResultType
value( 2 );
91
92
value[0] = 1.0 - std::exp(-4.0 * x( 0 )) * std::pow( std::sin(6 * M_PI * x( 0 ) ), 6);
93
94
double
mean
= sum(x) - x(0);
95
mean
/= (
numberOfVariables
() - 1.0);
96
97
double
g = 1.0 + 9.0 * std::pow(
mean
, 0.25);
98
double
h = 1.0 -
sqr
(value[0] / g);
99
value[1] = g*h;
100
101
return
value;
102
}
103
private
:
104
BoxConstraintHandler<SearchPointType>
m_handler;
105
};
106
107
}}
108
#endif