MATLAB: Fmincon: firstorderopt at each iteration

fmincon firstorderopt optimvaluesMATLAB

how to find the source code where optimValues.firstorderopt is computed?

optimValues.firstorderopt is shared by several iterative optimization algorithms. I want to see the formula used for it. But I can't trace it down, "Find Files" tried but no work.

Thank you!

Best Answer

I understand that you want to figure out how 'optimValues.firstorderopt' is computed. I executed a simple example of fmincon along with the a custom ouputput function. Then I placed a break point and stepped in through the different functions that were being called just before my custom output function. It appears that the 'firstorderopt' variable is set in the file 'nlconst.m' at line 952 or 954 inside the 'callOutputAndPlotFcns' to a variable called 'optimerror'. This variable 'optimerror' is set at line 407 and is based on 'normgradLag' and 'normcomp' variables computed at line 404 and 405 respectively of the same file 'nlconst.m'. Using breakpoints and stepping into the file would help find the formula of calculating the 'firstorderopt'.

You can edit the file 'nlconst.m' using the following command

>> edit nlconst.m

Hope this helps.

Related Solutions

MATLAB: How to change the formatting of the table displayed by FMINCON in the Optimization Toolbox 6.2

There is no way to directly set the format.

As a workaround, you will need to set 'Display' to 'final' and create a custom 'OutputFcn'. This function can use its input arguments to mimic the table generated by setting 'Display' to 'iter'.

Below is an example that mimics the table produced by FMINCON. Please change the code to meet your needs. If you want to prevent scientific formatting on numbers, you will probably want to change the "g"s to "f"s in the variable 'formatstr'. For example, change "%12.6g" to "%12.6f".

The variable 'formatstr' is coppied from NLCONST, the function called by FMINCON to perform the optimization. To make a similar table that is for another optimization function, you may need to search through the code in the Optimization Toolbox to find where the table is printed in that function.

A = [-1 -2 -2; ...
        1  2  2];
b = [0;72];
x0 = [10;10;10];    % Starting guess at the solution
   f = @(x)-x(1) * x(2) * x(3);
opt2 = optimset('Display','final','Algorithm','active-set','OutputFcn', @custom_output);
fmincon(f,x0,A,b,[],[],[],[],[],opt2);
function stop = custom_output(~, optimValues, state)
stop = false;
header = ['                                Max     Line search  Directional  First-order\n' ...
' Iter F-count        f(x)   constraint   steplength   derivative   optimality Procedure\n'];
formatstr = '%5.0f  %5.0f %12.6g %12.4g %12.3g %12.3g %12.3g %s  %s';
switch state
      case 'init'
          fprintf(header);
      case 'iter'
          iter = optimValues.iteration;
          funcCount = optimValues.funccount;
          f = optimValues.fval;
          con =  optimValues.constrviolation;
          laststep = optimValues.lssteplength;
          grad = optimValues.gradient;
          direcderiv = optimValues.directionalderivative;
          fo = optimValues.firstorderopt;
          hessUpdateMsg = optimValues.procedure;
         fprintf(formatstr,iter,funcCount,f,con,laststep,direcderiv,fo,hessUpdateMsg);
         fprintf('\n');
end

MATLAB: Error for ‘fmincon’ for global optimization

It's a case sensitivity issue. Instead of 'LB' and 'UB', pass in 'lb' and 'ub'.

Best Answer

Related Solutions

MATLAB: How to change the formatting of the table displayed by FMINCON in the Optimization Toolbox 6.2

MATLAB: Error for ‘fmincon’ for global optimization

Related Question