about gatool

조회 수: 4 (최근 30일)
Natalia
Natalia 2012년 1월 5일
Hello! I have a fitness functiion:
function z = my_fun(x)
Skorost8.input(1,1).mf(1,1).params=[x(1),x(2),x(3)];
Skorost8.input(1,1).mf(1,2).params=[x(4),x(5),x(6)];
Skorost8.input(1,1).mf(1,3).params=[x(7),x(8),x(9)];
Skorost8.input(1,1).mf(1,4).params=[x(10),x(11),x(12)];
Skorost8.input(1,2).mf(1,1).params=[x(13),x(14),x(15)];
Skorost8.input(1,2).mf(1,2).params=[x(16),x(17),x(18)];
Skorost8.input(1,2).mf(1,3).params=[x(19),x(20),x(21)];
load_system('petina.mdl');
sim('petina.mdl');
close_system('petina.mdl');
z=simout.signals.values;
And petina.mdl file, which it use. The mdl-file write to workspace a massive of values, which must be a minimal (z=simout.signals.values;)
When I try to use it with gatool I get:
>> x=ga(@my_fun,21);
??? Subscripted assignment dimension mismatch.
Error in ==> fcnvectorizer at 14
y(i,:) = feval(fun,(pop(i,:)));
Error in ==> makeState at 47
Score =
fcnvectorizer(state.Population(initScoreProvided+1:end,:),FitnessFcn,1,options.SerialUserFcn);
Error in ==> gaunc at 41
state = makeState(GenomeLength,FitnessFcn,Iterate,output.problemtype,options);
Error in ==> ga at 279
[x,fval,exitFlag,output,population,scores] = gaunc(FitnessFcn,nvars, ...
Caused by:
Failure in user-supplied fitness function evaluation. GA cannot continue.
How to use correctly gatool in this case? thank you very much!

채택된 답변

Sean de Wolski
Sean de Wolski 2012년 1월 5일
dbstop if error
Which will stop deep inside of ga(). Now put a break point in first line of your objective function. Run the feval line that causes the error and step through your fitness function to find the error and analyze the inputs.
  댓글 수: 5
Sean de Wolski
Sean de Wolski 2012년 1월 5일
Yes, a 1x450 would be 450 objectives (which sounds intuitively like a lot (way too many) to me!). Especially since you only want to optimize 21 variables.
If you look in the doc for ga() on line one it explains the output:
"x = ga(fitnessfcn,nvars) finds a local unconstrained minimum, x, to the objective function, fitnessfcn. nvars is the dimension (number of design variables) of fitnessfcn. The objective function, fitnessfcn, accepts a vector x of size 1-by-nvars, and returns a scalar evaluated at x."
Could you explain what you're trying to optimize?
Natalia
Natalia 2012년 1월 5일
I try to optimize the parametrs of fuzzy-logic controller, which describe into the fis file and use in my mdl. gamultiobj has worked. I don't have a erorrs :) Thank you :)

댓글을 달려면 로그인하십시오.

추가 답변 (0개)

카테고리

Help CenterFile Exchange에서 Linear Least Squares에 대해 자세히 알아보기

제품

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by