I presently have this code which tests a paricular function which I created _ _ [y n] function_gen1pdf1(a,b,c) _ _ for N times. Now this works fine up to N=100. However, I want it to work for N=100,000 but it's taking much too long to compute. Is it possible to vectorize it or make it work faster? Any suggestions are much appreciated!
N=100000;
y=zeros(N,1);
n=zeros(N,1);
for i=1:N; %testing the function for 100000 times
[y(i) n(i)]= function_gen1pdf1(2,3,0.5); %The function being tested
end

7 Kommentare

Star Strider
Star Strider am 7 Feb. 2015
Please post the function code!
Stephen23
Stephen23 am 7 Feb. 2015
"Is it possible to vectorize it": this depends entirely on the function function_gen1pdf1. If you want help trying to vectorize this, then you have to upload its code too.
Geoff Hayes
Geoff Hayes am 7 Feb. 2015
rox - I don't understand the line
[y(i) n(i)]= function_gen1pdf1(2,3,0.5);
You keep calling function_gen1pdf1 with the same set of inputs for every i. Why is the code repeating the same call at every iteration of the for loop?
Tapan
Tapan am 7 Feb. 2015
I will echo everyone else's comments - you need to provide the function code. Perhaps try a parlor loop in the meantime.
Jan
Jan am 7 Feb. 2015
@Geoff: The "gen" in the name might mean a genetic algorithm, such that random effects can be the reasn for different results inspite of the same inputs.
Image Analyst
Image Analyst am 7 Feb. 2015
Or maybe it generates a PDF document, or generates a probability distribution function. Whatever it does, apparently PDF does not mean "Pretty Darn Fast".
Jan
Jan am 7 Feb. 2015
Bearbeitet: Jan am 7 Feb. 2015
@rox: While the "function_" in the name of your function is pretty meaningless, "gen1pdf1" catches our interest.

Melden Sie sich an, um zu kommentieren.

 Akzeptierte Antwort

Image Analyst
Image Analyst am 7 Feb. 2015

0 Stimmen

It's your function, it's not the for loop. If you just run a for loop 100,000 times it takes 0.2 milliseconds.
I tried it:
tic
for k = 1 : 100000
;
end
toc
Elapsed time is 0.000219 seconds.
So it if takes a long time, it's due to your function_gen1pdf1() function.

3 Kommentare

rox
rox am 7 Feb. 2015
I guess the problem is that function_gen1pdf1(2,3,0.5) calls yet another function inside it (as it is based on the rejection method) which contains fplot. Now I was experimenting and just tried working it out without the fplot, and it's much much faster! Thanks for your clarifying that the problem was with the function itself!
Image Analyst
Image Analyst am 7 Feb. 2015
rox, you're welcome. Does that mean that you can now mark the answer as "Accepted"?
rox
rox am 7 Feb. 2015
Yes, thanks!

Melden Sie sich an, um zu kommentieren.

Weitere Antworten (0)

Kategorien

Mehr zu Loops and Conditional Statements finden Sie in Hilfe-Center und File Exchange

Gefragt:

rox
am 7 Feb. 2015

Bearbeitet:

Jan
am 7 Feb. 2015

Community Treasure Hunt

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

Start Hunting!

Translated by