how to run function for several times with different variables

18 vues (au cours des 30 derniers jours)
Asliddin Komilov
Asliddin Komilov le 5 Mar 2019
Commenté : Stephen23 le 12 Mar 2019
I have something like this:
a=5
b=8
c=9
d=11
%a,b,c and d are function inputs
n=a*0.1:a*0.1:a
for i=1:length(n)
"function"
end
but I need to run function replacing 'a' in 'n' with 'b','c' and 'd' at once and save the data so I can compare it. it is a sensitivity analysis for this function outputs.
  2 commentaires
dpb
dpb le 5 Mar 2019
To do every point for all four variables would be quad-nested for loops and you'd end up with 4^10= 1048576 outputs. Depending on how expensive function() is in compute time, this might take a while...
Typically one uses response surface methods or similar to do such studies that replace the detail function with a (typically) quadratic surface over the design space. (Myers, RH, Response Surface Methodology, Allyn and Bacon, 1971 was my "go to" and still a favorite).
Asliddin Komilov
Asliddin Komilov le 12 Mar 2019
that is too difficult for me, I did it for each variable separately (see below). thank you

Connectez-vous pour commenter.

Réponses (2)

Dennis
Dennis le 5 Mar 2019
In your example you do not need more than one variable:
a=[5 8 9 11];
for i=1:numel(a)
n=a(i)*0.1:a(i)*0.1:a(i);
for k=1:numel(n)
%function
end
end
  2 commentaires
dpb
dpb le 5 Mar 2019
"but I need to run function replacing 'a' in 'n' with 'b','c' and 'd'..."
Asliddin Komilov
Asliddin Komilov le 12 Mar 2019
thanks, but it gave error inside th function, so I decided to skip it.

Connectez-vous pour commenter.


Asliddin Komilov
Asliddin Komilov le 12 Mar 2019
Modifié(e) : Stephen23 le 12 Mar 2019
I just saved the loop for each variable.
a=5
b=8
c=9
d=11
%a,b,c and d are function inputs
n=a*0.1:a*0.1:a
for i=1:length(n)
"function"
end
save file_a
a=a(end);
clear "function outputs"
%--------------------------
n=b*0.1:b*0.1:b
for i=1:length(n)
"function"
end
save file_b
b=b(end);
clear "function outputs"
%---------------------------
..........
  1 commentaire
Stephen23
Stephen23 le 12 Mar 2019
You could save the bother by simply putting your data into one array and using indexing.

Connectez-vous pour commenter.

Catégories

En savoir plus sur Get Started with MATLAB dans Help Center et File Exchange

Community Treasure Hunt

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

Start Hunting!

Translated by