Can you help me with the output of my optimizer?

1 Ansicht (letzte 30 Tage)
Jorge Armando Vazquez
Jorge Armando Vazquez am 1 Jun. 2021
Bearbeitet: Walter Roberson am 3 Jun. 2021
I have this program from
par0 = [0.1,0.0001,100];% initial values
history = [];
fun_objetivo = @(par,pfrac)FunObjetivo(par);
options = optimset('MaxIter',3000,'MaxFunEvals',3000,'Display','iter','OutputFcn', @myoutput);
%We use here the Nelder-Mead method
par_optimos = fminsearch(fun_objetivo, par0, options);
function stop = myoutput(par_optimos,optimvalues,state);
stop = false;
if isequal(state,'iter')
history = [history;par_optimos];
end
end
but I can´t make it give me the value of my vector par_optimos for each iteration, can someone help me?

Antworten (2)

Sulaymon Eshkabilov
Sulaymon Eshkabilov am 1 Jun. 2021
You'd need define a math formulation of fun_objetivo = @(par,pfrac)FunObjetivo(par);
Once it is done, you can run via loop iteration, i.e.: for ... end

Alan Weiss
Alan Weiss am 2 Jun. 2021
Use this example as a guide: Example of a Nested Output Function.
Alan Weiss
MATLAB mathematical toolbox documentation
  2 Kommentare
Jorge Armando Vazquez
Jorge Armando Vazquez am 3 Jun. 2021
That is what I used
fun_objetivo = @(par,pfrac)FunObjetivo(par);
history=[];
options = optimset('MaxIter',3000,'MaxFunEvals',3000,'Display','iter','OutputFcn', @myoutput);;%,'PlotFcns',@optimplotfval);
%We use here the Nelder-Mead method
par_optimos = fminsearch(fun_objetivo, par0, options);
function stop = myoutput(par_optimos,optimvalues,state);
stop = false;
if isequal(state,'iter')
history = [history;par_optimos];
end
end
but there is error with
Error in Main_optim_15_05_21>myoutput (line 15)
history = [history;par_optimos];
Walter Roberson
Walter Roberson am 3 Jun. 2021
Bearbeitet: Walter Roberson am 3 Jun. 2021
You did not nest the function like suggested.
history = do_the_work;
function hist = do_the_work()
fun_objetivo = @(par,pfrac)FunObjetivo(par);
history=[];
options = optimset('MaxIter',3000,'MaxFunEvals',3000,'Display','iter','OutputFcn', @myoutput);;%,'PlotFcns',@optimplotfval);
%We use here the Nelder-Mead method
par_optimos = fminsearch(fun_objetivo, par0, options);
hist = history;
function stop = myoutput(par_optimos,optimvalues,state);
stop = false;
if isequal(state,'iter')
history = [history;par_optimos];
end
end
end

Melden Sie sich an, um zu kommentieren.

Kategorien

Mehr zu Optimization finden Sie in Help Center und File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by