[galib] ex26 of GALIB

[assia]

Hi every body,

Please, I need some hleps! 

I'm trying to use ex26 for my problem's optimization. I adapted it for my needs. I create a new Class of objects, and converted all ex26's functions for using my object insteed integer. All things are OK  objective, mutator, intializer and crossover  functions.  

My problem, I think  is in Garandom, I found the same result each time. I did some modifcations in seed, but I had the same results. The program take randomly the object in the same area, by the bounds (413 in my exemple in files with this mail).

 with tihs mail, I join two exemples of results I gererate with my program.

Could you help me please to understand this problem and find a good result.
If you need any other nformations just mail me.

Best regards.
PS: I'm sorry for my bad english...
Assia.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/pipermail/galib/attachments/20070309/56b85f97/attachment.htm
-------------- next part --------------
120	# current generation
1.01031	# current convergence
120000	# number of selections since initialization
120000	# number of crossovers since initialization
120000	# number of mutations since initialization
120000	# number of replacements since initialization
121000	# number of genome evaluations since initialization
121	# number of population evaluations since initialization
105.832	# maximum score since initialization
5.82167	# minimum score since initialization
9.72493	# average of all scores ('on-line' performance)
10.6528	# average of maximum scores ('off-line' performance)
7.07265	# average of minimum scores ('off-line' performance)

56.0019	# mean score in initial population
105.832	# maximum score in initial population
20.8857	# minimum score in initial population
14.4207	# standard deviation of initial population
-1	# diversity of initial population (0=identical,-1=unset)

6.66611	# mean score in current population
6.91968	# maximum score in current population
5.88167	# minimum score in current population
0.210635	# standard deviation of current population
-1	# diversity of current population (0=identical,-1=unset)

20	# how far back to look for convergence
100	# how often to record scores
0	# how often to write scores to file
generations.dat	# name of file to which scores are written

Rect       r   c  nr  nc  Ef     Num    Prob
--------------------------------------------
Rect   0   0   2   1   1   1      95    1.78
Rect   1   3   4   1   1   2     439    1.65
Rect   2   3   4   1   1   3     439    1.65
Rect   3   1   0   1   1   0     206    1.65
Rect   4   3   0   1   1   5     413    1.43
Rect   5   2   0   1   1   6     332    1.22
Rect   6   3   4   1   1   4     439    1.65
--------------------------------------------
                          Sum Prob     11.03
                          Fitness    9.65088
                          Score      5.82167

-------------- next part --------------
999	# current generation
1	# current convergence
3996000	# number of selections since initialization
3996000	# number of crossovers since initialization
3996000	# number of mutations since initialization
3996000	# number of replacements since initialization
4000000	# number of genome evaluations since initialization
1000	# number of population evaluations since initialization
119.282	# maximum score since initialization
5.51167	# minimum score since initialization
6.05436	# average of all scores ('on-line' performance)
6.19973	# average of maximum scores ('off-line' performance)
5.641	# average of minimum scores ('off-line' performance)

56.3679	# mean score in initial population
119.282	# maximum score in initial population
16.2817	# minimum score in initial population
14.6929	# standard deviation of initial population
-1	# diversity of initial population (0=identical,-1=unset)

5.51172	# mean score in current population
5.51167	# maximum score in current population
5.51167	# minimum score in current population
4.98182e-005	# standard deviation of current population
-1	# diversity of current population (0=identical,-1=unset)

20	# how far back to look for convergence
100	# how often to record scores
0	# how often to write scores to file
generations.dat	# name of file to which scores are written

Rect       r   c  nr  nc  Ef     Num    Prob
--------------------------------------------
Rect   0   0   1   1   1   1      67    1.65
Rect   1   3   0   1   1   6     413    1.13
Rect   2   3   0   1   1   3     413    1.63
Rect   3   3   0   1   1   5     413    1.43
Rect   4   3   0   1   1   4     413    1.58
Rect   5   1   0   1   1   0     206    1.65
Rect   6   3   4   1   1   2     439    1.65
--------------------------------------------
                          Sum Prob     10.72
                          Fitness    9.09427
                          Score      5.51167