How to fix: slow loops in R

Question

I'm facing some trouble with this R looping. I imagine it's because R isn't so prepared to code this kind of looping.

Thank you in advance!

df <- data.frame("x1" = NULL)

a <- 1
j <- 1

while (a<201)
{
  for(i in 1:1156)
  {
    df[j,"x1"] <- j
    j <- j + 1
  }
  a <- a + 1
}

length(df$pontos)
df

add warnigs or error or output of code if any to help solve this better. — Devidas, Mar 26 '19 at 12:55
In my opinion the code is equivalent to `j <- 200*1156; df <- data.frame(x1=1:j); a <- 201; j <- j+1` — jogo, Mar 26 '19 at 13:04
Possible duplicate of [Why are loops slow in R?](https://stackoverflow.com/questions/7142767/why-are-loops-slow-in-r) — divibisan, Apr 02 '19 at 17:47

score 2 · Answer 1 · answered Mar 26 '19 at 13:03

2

There are no errors in the codes actually. If you put print(a) after a <- a + 1, you can see that the codes work but extremely slow. I think that is the reason why you thought the loop never ends.

answered Mar 26 '19 at 13:03

Bowen

95
11

Yes, there are no erros but the code is extremely slow. – Felipe Dourado Mar 26 '19 at 13:15
1

@FelipeDourado You seem to create one-column data frame from 1 to 200 * 1156. As a more efficient method, you can just pass the `1:(200 * 1156)` as the argument to `data.frame`. – Bowen Mar 26 '19 at 13:19

score 0 · Accepted Answer · answered Mar 26 '19 at 13:07

0

It runs for me if you decrease the iterations in the inner for loop to 100. It might just be taking a while. Could the outer loop just be a for loop as well?

answered Mar 26 '19 at 13:07

Graeme Prentice-Mott

302
1
10

If the outer loop just be a for loop, how would it help? My real problem is the delay to run the code. – Felipe Dourado Mar 26 '19 at 13:14

How to fix: slow loops in R

2 Answers2