Problem with rng shuffle
53 views (last 30 days)
Aftab Ahmed Khan on 25 Feb 2015
Hello everyone, I am running my simulation for like 50000 iterations but when i use
during each iteration, the processing speed of my computer gets very slow. I want to understand that why this one line of code has so much effect on my computer ?
Peter Perkins on 25 Feb 2015
As Sean said, don't reset the random number generator in a loop. There's no good reason to do it. You are NOT making your random number "more random", and in fact, you are likely making them less random in very complicated ways. DON'T DO THIS. I can't say that strongly enough.
'shuffle' pauses for .01s each time you call it to ensure that it uses a new seed that's different than the last time you called it. That's why it's slowing down your code.
More Answers (2)
Sean de Wolski on 25 Feb 2015
Why do you want to reset the stream on each iteration? This will be slow. The stream is random, so just set it once at the beginning of the run.
Chuck on 5 May 2016
I have noticed this too. Even if you do not shuffle in every generation and shuffle it once, it will be slower than rng("default"). Of course, it might be unnoticeable depending on your code.
But... We do not know the nature of your iterations. If you are using parallelization (e.g. parfor), than you need to do rng("shuffle") right after the parfor definition line. Otherwise, after each individual for loop is done, it goes back go rng("default").
If your iterations are not using parallelization, then you should be fine by just adding that at the beginning of your code.