How can I select 20% of a given data set for training?

I have a data set that contains 2000 data points

and I would like to select (randomly) 20% of the data for training. How would i do that?

And suppose i wanted the percentage to be a variable? So i can sample any percentage i define?

Thank you!

Answer 1

[Dnyanesh Suryavanshee]

data = 2000;

per = 20;

ans=data(per/100);

Answer 2

[Rik]

If you want the sample to be repeatable, you will need to specify the state of the random number generator by calling the rng function.

p = randperm(n,k); %returns a row vector containing k unique integers selected randomly from 1 to n.

To calculate k as a function of n and the percentage, you can use the normal arithmatic:

k=round(n*perc/100);

You need to round (or use floor or ceil), because the inputs to randperm must be integer.