[R] How to speed this up?

[Thierry Onkelinx]

Dear Jen,

Vectorisation is the keyword here. 250k sample takes only 2.5 seconds on my
machine. 2.5 million takes 29 seconds.

n <- 250e3

# country code
cc <- "+212"
# prefixes
IAM <- c(610,        611,    613,    615,    616,
618,    641,    642,    648,    650,    651,    652,    653,
          654,  655,    658,    659,    661,    662,    666,    667,
668,    670,    671,    672,    673,
          676,  677,    678)
Medi <- c(612,        614,    617,    619,    644,
645,    649,    656,    657,    660,    663,    664,    665,
          669,  674,    675,    679)
MOROC <- c(0636, 0637)
prefix <- rbind(
  data.frame(
    region = "IAM",
    prefix = IAM
  ),
  data.frame(
    region = "Medi",
    prefix = Medi
  ),
  data.frame(
    region = "MOROC",
    prefix = MOROC
  )
)
prefix <- merge(
  prefix,
  as.data.frame(table(region = prefix$region))
)
system.time({
  prefix.sample <- sample(prefix$prefix, n, prob = prefix$Freq, replace =
TRUE)
  nums <- apply(
    matrix(
      sample(0:9, 6 * n, replace = TRUE),
      ncol = 6
    ),
    1,
    paste,
    collapse = ""
  )
  phonenumbers <- paste0(cc, prefix.sample, nums)
})


ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and
Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium

To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to say
what the experiment died of. ~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data. ~ Roger Brinner
The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey

2017-02-28 17:22 GMT+01:00 Jen <plessthanpointohfive op gmail.com>:

> Hi, I'm trying to generate 2.5 million phone numbers.  The code below
> generates a random sample of 250K MPNS for Morocco.  It takes about 10
> minutes.
>
> I need to generate 2.5 million.  I've run it through once and it took about
> 45 hours.
>
> Is there a way to speed this up?
>
> Thanks,
>
> Jen
>
> # generate random sample of mobile phone numbers (MPNs) - Morocco
>
> # Mobile phone number format:  +212-6xx-xxxxxx
>
> library(data.table)
>
> # country code
>
> cc <- "+212"
>
> # prefixes
>
> IAM  <- data.table(matrix(c(610,        611,    613,    615,    616,
> 618,    641,    642,    648,    650,    651,    652,    653,
>           654,  655,    658,    659,    661,    662,    666,    667,
> 668,    670,    671,    672,    673,
>           676,  677,    678), dimnames=list(NULL, "IAM")))
>
>
>
> Medi <- data.table(matrix(c(612,        614,    617,    619,    644,
> 645,    649,    656,    657,    660,    663,    664,    665,
>           669,  674,    675,    679), dimnames=list(NULL, "Medi")))
>
> MOROC <- data.table(matrix(c(0636, 0637), dimnames=list(NULL, "MOROC")))
>
> # combine
>
> mno <- c(IAM, Medi, MOROC)
>
> # generate MPNs
> MPN <- NULL
>
> system.time(for (i in 1:250000){
> # randomly select number from list
>
> prefix <- sapply(mno[floor(runif(1, 1, length(mno)+1))], function(x)
> sample(x, 1))
>
> MNO <- names(prefix)
>
> # randomly generate 6 numbers between 0 and 9, inclusive
>
> nums <- floor(runif(6, 0, 9))
>
> # concatenate
>
> tmp <- c(paste(c(cc,prefix,t(nums)), sep="", collapse=""), MNO)
>
> MPN[[i]] <- tmp
>
> i <- i+1
>
>
> })
>
> # unlist
>
> df <- data.table(matrix(unlist(MPN), nrow=length(MPN), ncol=2, byrow=T,
> dimnames = list(seq(1, length(MPN),1), c("MPN", "MNO"))   ))
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help op r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/
> posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

	[[alternative HTML version deleted]]