eMOEA
所属分类:数学计算
开发工具:C/C++
文件大小:15KB
下载次数:4
上传日期:2011-04-19 11:25:04
上 传 者:
ttdbk
说明: an epsilon Multi-Objective EA
文件列表:
eMOEA\archive.out (2744, 2004-01-04)
eMOEA\ar_variables.out (6272, 2004-01-04)
eMOEA\box_vectors.out (2744, 2004-01-04)
eMOEA\eps-dom.cpp (17059, 2004-01-04)
eMOEA\inp (57, 2003-04-22)
eMOEA\input.h (1096, 2003-04-22)
eMOEA\popltn.out (2800, 2004-01-04)
eMOEA\random.h (3908, 2003-04-18)
eMOEA\realcross.h (2444, 2003-04-20)
eMOEA\realmut.h (1516, 2003-04-20)
eMOEA (0, 2011-04-19)
------------------------------------------------------------------------
========================================================================
R E A D M E F I L E
========================================================================
The complete package consists of the following files :
-eps-dom.cpp
-inp
-input.h
-random.h
-Readme (this file)
-realcross.h
-realmut.h
The small descritption about these files are given below:
eps-dom.cpp
~~~~~~~~~~~
This file consists of the main code for the implementation of the Epsilon
Based MOEA propsed by Dr.Deb et. al. in the paper titled "Towards a Quick
Computation of Well Spread Pareto-Optimal Solutions"
- Deb, Mohan and Mishra (EMO-2003 Proceedings)
Also available from http://www.iitk.ac.in/kangal/pub.htm
(KanGAL Report Number 2003002)
This file has most of the functions (including the main() function). The number
of objective functions, the number of decision variables, the number of
constraints ,the actual function definitions and the actual constraints are defined
inside this file. The epsilon matrix is also defined in this file.
NOTE: (1) The dimension of the epsilon matrix is (1xM)
where M = No. of Objective functions
(2) The code is written for minimization cases.
For maximization, transform the
problem into a minimization problem by multiplying by (-1).
realcross.h
~~~~~~~~~~~
Contains the implementation of the real parameter SBX crossover. For details,
please refer to Deb's 2001 book entitled `Multi-Objective Optimization
using Evolutionary Algorithms', Chichester: Wiley.
realmut.h
~~~~~~~~~
Contains the implementation of polynomial mutation. For details, please
refer the above-mentioned book.
input.h
~~~~~~~
Contains the function which takes the upper/lower bounds corresponding to each
decision variable. It also prompts for the GA parameters.
random.h
~~~~~~~~
This file has an starts up the random number generator based on a seed value
(which is provided as an input)
inp
~~~
A Sample input file. The current file corresponds to the 7-decision variable,
3-objective, 0-constraint DTLZ1 function.
Refer the following paper for complete and detailed description of the
problem:
Scalable multi-objective optimization test problems,
- K. Deb and L. Thiele and M. Laumanns and E. Zitzler
http://www.iitk.ac.in/kangal/pub.htm (Report No. 2001001) }
-------------- Sample Input File --------------------
0 1 // lower and upper bound of first decision variable
0 1
0 1
0 1
0 1
0 1
0 1 // lower and upper bound of Seventh decision variable
1.0 // Probability of crossover (keep 1.0)
0.1428 // Probability of mutation (keep 1/number of variables)
15 // SBX distribution index (keep between 5 to 20)
20 // Mutation distriubution index (keep between 10 to 100)
0.123 // Random seed (keep between 0.0 and 1.0)
15000 // Number of generations (note that number of function eval.
// is TWICE to this number)
-----------------End of Sample Input File ------------------
H O W T O U S E T H E C O D E
------------------------------------
The program is complied by the command (requires C++ complier)
c++ eps-dom.cpp
It can be executed by giving:
a.out OR ./a.out
The inputs can be given from the console or can be redirected from
a file like:
a.out
近期下载者:
相关文件:
收藏者: