Minimizing a function using the BFGS method
Ältere Kommentare anzeigen
I can not minimize it using the fminunc function. While respecting that both alpha and beta values are between 0 and 1. Thanks!
Antworten (1)
John D'Errico
am 7 Jul. 2016
Bearbeitet: John D'Errico
am 7 Jul. 2016
0 Stimmen
You cannot do so using BFGS either. However, you COULD use FMINCON to do so. And if you have fminunc, then you have fmincon.
So what is the point of this question? In fact, you have no question here that I see. Writing your own tools for something when you already have professionally written tools to do the same thing is just a waste of time.
8 Kommentare
amine&&
am 7 Jul. 2016
Torsten
am 7 Jul. 2016
x = fmincon(@(x)mFunction(x(1),x(2)),x0,A,b,Aeq,beq)
Best wishes
Torsten.
John D'Errico
am 7 Jul. 2016
That just means you did not use fmincon properly, as shown how by Torsten.
amine&&
am 7 Jul. 2016
amine&&
am 7 Jul. 2016
John D'Errico
am 7 Jul. 2016
Torsten showed you how to do it.
You made a mistake in not formulating your function properly. First, an optimizer passes in a vector of unknowns. Your function has TWO arguments, two unknowns. So Torsten showed you that the function needs to be passed its arguments properly.
Next, when you call it like this:
x = fmincon(mFunction,x0,A,b,Aeq,beq)
MATLAB does not understand that you want to pass a function to fmincon. It sees the function and tries to evaluate it instead. So Torsten showed you how to use a function handle.
amine&&
am 7 Jul. 2016
Kategorien
Mehr zu Solver Outputs and Iterative Display finden Sie in Hilfe-Center und File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!