CODEWITHSUNDEEP
r
Pranav
Pranav

Reputation: 3312

A good way to create a sequence of pairs in R?

I want to create a sequence of random 2D points in R using runif. What is a good way to do that?

This is my current implementation(does not work)

getpoints <- function(n)
  {
    t <- 1:n;
    for (i in 1:n)
      {
        t[i] <- runif(2,-1,1)
      }
    return (t)
  }

Upvotes: 1

Views: 1043

Answers (4)

sebastian-c
sebastian-c

Reputation: 15415

Just adding to the numerous answers you have here. This one, like @Arun's is easily able to handle more than 2 columns if necessary.:

replicate(2, runif(100))

It may almost be benchmarking time.

Upvotes: 2

PascalVKooten
PascalVKooten

Reputation: 21481

To give an overview of this thread, first of all I will show the speeds associated with the several methods used (so you can draw your own conclusion about which is the best), then I will show you how you could use a function; i.e. the way you intended your function probably.

n <- 1000000

> system.time(cbind(runif(n, -1, 1), runif(n, -1, 1)))
   user  system elapsed 
   0.28    0.03    0.31 
> system.time(matrix(runif(2*n, min = -1, max = 1), ncol = 2))
   user  system elapsed 
   0.32    0.00    0.33 
> system.time(replicate(2, runif(n, -1, 1)))
   user  system elapsed 
   0.26    0.02    0.35

Here is the function, and the time benchmark:

getPoints <- function (N) {
  pointMatrix <- matrix(1, nrow = 1000000, ncol = 2) 
  for (i in 1:N) {
    pointMatrix[i, ] <- runif(2, -1, 1) # it needs to run for all columns
  }
}

> system.time(getPoints(n))
   user  system elapsed 
  17.59    0.04   19.94 
>

Upvotes: -1

Arun
Arun

Reputation: 118889

n <- 200
o <- matrix(runif(2*n, min = -1, max = 1), ncol = 2)

Upvotes: 9

agstudy
agstudy

Reputation: 121618

 n <- 5
 cbind(runif(n,-1,1),runif(n,-1,1))
            [,1]       [,2]
[1,] -0.68434317 -0.7772889
[2,] -0.91200792 -0.6408075
[3,] -0.01888610  0.7350491
[4,]  0.01782097  0.9674700
[5,] -0.56707264 -0.9991566

Upvotes: 3

Related Questions