MATLAB Answers

Invalid problem structure error when using GlobalSearch

9 views (last 30 days)
Alexandra
Alexandra on 17 Oct 2013
Commented: Alexandra on 17 Oct 2013
When I use lsqcurvefit(problem) & run(multistart,problem) - I get no errors with my problem structure but now that I am attempting GlobalSearch I get an error message:
Error using AbstractGlobalSolver/checkProblem (line 227)
Invalid problem structure: missing required field or invalid field value.
Error in GlobalSearch/run (line 305)
obj.checkProblem(problem, probRequiredFields, probValidValues);
My code is as follows:
xrise = xdata(1:300);
yrise = ydata(1:300);
options = optimoptions('lsqcurvefit');
TolFun_Data = 1e-4;
TolX_Data= 1e-4;
FinDiffRelStep_Data=[0.01];
MaxIter_Data = 1e5
options = optimoptions(options,'Display', 'final');
options = optimoptions(options,'MaxIter', MaxIter_Data);
options = optimoptions(options,'TolFun', TolFun_Data);
options = optimoptions(options,'TolX', TolX_Data);
options = optimoptions(options,'FunValCheck', 'off');
options = optimoptions(options,'Algorithm', 'trust-region-reflective');
options = optimoptions(options,'Diagnostics', 'off');
options = optimoptions(options,'FinDiffRelStep', FinDiffRelStep_Data);
options = optimoptions(options,'FinDiffType', 'central');
options = optimoptions(options,'Jacobian', 'off');
options = optimoptions(options,'JacobPattern', 'sparse(ones(jrows,jcols))');
options = optimoptions(options,'TypicalX', 'ones(numberofvariables,1)');
x0 = [0.05];
lb = [0.05];
ub = [0.3];
problem =createOptimProblem('lsqcurvefit','x0',x0,...
'objective',@myfuncTi,'xdata',xrise,'ydata',...
yrise,'lb',lb,'ub',ub,'options',options);
gs=GlobalSearch;
[TiVal fval exitflag output solutions]=run(gs,problem)

  0 Comments

Sign in to comment.

Accepted Answer

Alan Weiss
Alan Weiss on 17 Oct 2013
GlobalSearch does not accept lsqcurvefit as a local solver. If you want to use GlobalSearch, your only choice of local solver is fmincon.
I think that error message could be improved. I will look into it.
Alan Weiss
MATLAB mathematical toolbox documentation

  1 Comment

Alexandra
Alexandra on 17 Oct 2013
Thank you for your response. I agree that can definitely be made clearer.

Sign in to comment.

More Answers (0)

Community Treasure Hunt

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

Start Hunting!

Translated by