while and for in R

Question

I want to creat an alogrithm in R using while and for such that, given a vector V and an integer M, sums the elements of the vector V until that sum exceeds M, and I want it to inform the value that exceeded M as well as how many repetitions it took in order to exceed such number (i.e., how many elements of the vector did it summed).

It's easier to help you if you provide a [reproducible example](https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example) with sample input and desired output that can be used to test and verify possible solutions. — MrFlick, Aug 25 '22 at 20:36
You're right. Say: V = [1 2 3 4 5]; X = 7; V[1] + V[2] = 1 + 2 = 3 <= 7; V[1] + V[2] + V[3] = 1 + 2 +3 = 6 <= 7; V[1] + V[2] + V[3] + V[4] = 1 + 2 + 3 + 4 = 10 > 7; So I want the algorithm to stop at 4, inform the value of 10, and tell me the number of iterations it took (again, in this case, 4). Hope it's clear! — Gaspar, Aug 25 '22 at 20:37
I used X instead of M here, but it's the same. It's the value it has to exceed. — Gaspar, Aug 25 '22 at 20:44

score 2 · Accepted Answer · answered Aug 25 '22 at 21:14

Here are three functions.

fun uses vectorised cumsum to sum the vector;
The other two have self-descriptive names.

They all return a named vector with the iterations count and the first sum exceeding m.

fun <- function(v, m) {
  i <- which(cumsum(v) < m)
  if(length(i) + 1L < length(v))
    i <- c(i, length(i) + 1L)
  c(Iter = length(i), Sum = sum(v[i]))
}
fun_while <- function(v, m) {
  Sum <- 0L
  Iter <- 0L
  while(Iter < length(v)) {
    Iter <- Iter + 1L
    Sum <- Sum + v[Iter]
    if(Sum > m) break
  }
  c(Iter = Iter, Sum = Sum)
}
fun_for <- function(v, m) {
  Sum <- 0L
  for(i in seq_along(v)) {
    Sum <- Sum + v[i]
    if(Sum > m) break
  }
  if(is.null(i))   i <- 0L
  c(Iter = i, Sum = Sum)
}

V <- 1:5
M <- 7
fun(V, M)
#> Iter  Sum 
#>    4   10
fun_while(V, M)
#> Iter  Sum 
#>    4   10
fun_for(V, M)
#> Iter  Sum 
#>    4   10

V <- 1:5
M <- 20
fun(V, M)
#> Iter  Sum 
#>    5   15
fun_while(V, M)
#> Iter  Sum 
#>    5   15
fun_for(V, M)
#> Iter  Sum 
#>    5   15

V <- 1:10
M <- 42
fun(V, M)
#> Iter  Sum 
#>    9   45
fun_while(V, M)
#> Iter  Sum 
#>    9   45
fun_for(V, M)
#> Iter  Sum 
#>    9   45

V <- NULL
M <- 7
fun(V, M)
#> Iter  Sum 
#>    0    0
fun_while(V, M)
#> Iter  Sum 
#>    0    0
fun_for(V, M)
#> Iter  Sum 
#>    0    0

^{Created on 2022-08-25 by the reprex package (v2.0.1)}

I was comparing your fun_for vs my f.loop, no clue why, but the max of your function is high: Unit: microseconds expr min lq mean median uq max neval cld f.loop(v, val) 77.6 82.50 96.6310 85.20 88.00 440.5 1000 a fun_for(v, val) 70.7 75.20 93.0540 77.50 80.65 5389.6 1000 a (ie 440 vs 5389) — Will, Aug 25 '22 at 21:35
Thank you very much, truly. But in reality I am looking for something much, much simpler. Using just ```for``` and ```while```. If somebody comes up with an idea for that, I'd be truly appreciated. — Gaspar, Aug 25 '22 at 22:04
@Gaspar I don't understand this last comment, one of the functions uses only `while` and another uses only `for`. — Rui Barradas, Aug 26 '22 at 04:42

Will · Answer 2 · 2022-08-25T21:22:44.817

1

both looping and vectorial way. benchmark added for curiosity.

library(data.table)
library(microbenchmark)
library(ggplot2)

val <- 2e6
v <- seq(1, 1e6, 2)

f.vec <- function(v, val) {
  dt <- as.data.table(v)
  dt[, cs := cumsum(v)]
  idx <- dt[cs <= val, .N]
  valnext <- dt[idx+1, .(v)]
  list(rep = idx, valnext = valnext)
}

f.loop <- function(v, val) {
  idx <- 1
  total <- 0
  for (el in v) {
    total <- total + el
    if (total >= val) {
      break
    } 
    idx <- idx + 1
  }
  list(rep = idx-1, valnext = v[idx])
}

f.vec(v, val)
#> $rep
#> [1] 1414
#> 
#> $valnext
#>       v
#> 1: 2829
f.loop(v, val)
#> $rep
#> [1] 1414
#> 
#> $valnext
#> [1] 2829

tm <- microbenchmark(f.vec(v, val), f.loop(v, val), times=1000L)
autoplot(tm)
#> Coordinate system already present. Adding new coordinate system, which will replace the existing one.

tm
#> Unit: microseconds
#>            expr    min     lq      mean median     uq     max neval cld
#>   f.vec(v, val) 4848.5 5163.9 7856.0267 5643.6 9146.1 66502.9  1000   b
#>  f.loop(v, val)   72.7   75.8   81.1795   79.1   81.2   255.2  1000  a

^{Created on 2022-08-25 with reprex v2.0.2}

edited Aug 25 '22 at 21:22

answered Aug 25 '22 at 20:51

Will

910
7
17

Thank you very much for your answer, but I need to do it using the while function! – Gaspar Aug 25 '22 at 20:52
@Gaspar Why do you need to do it using `while`, is it homework? If so, please edit the question with that information, not even in the question title you say that `while` is *needed*. – Rui Barradas Aug 25 '22 at 20:56
It's the syntax I use. And the title does say the name of the functions "while" and "for". But will make it clearer anyways. – Gaspar Aug 25 '22 at 21:00
2

Gaspar> you're right, I ignored your requirements because I assumed it was a beginner mistake. R is a vectorial language. it does not make sense to do it with a loop. and the performance will kill you. but added with a loop to answer your question. – Will Aug 25 '22 at 21:07

while and for in R

2 Answers2