HI Paul,
I had a problem a bit like this when trying to implement the Lim-Wolfe
imputation for ranks. The size of the output matrix increases
exponentially with the size of the initial matrix and it goes into
disk-swapping slow motion. It works for small matrices, but rapidly
runs out of memory. I have planned to insert code to only retain the
largest maximum to fix this. If I get to it, I'll post the result.
Jim
On Sun, Apr 24, 2022 at 11:01 AM Paul Bernal wrote:
>
> Dear R friends,
>
> Hope you are doing great. The reason why I am contacting you all, is
> because the code I am sharing with you takes forever. It started running at
> 2:00 AM today, and it's 7:52 PM and is still running (see code at the end
> of this mail).
>
> I am using Rx64 4.1.2, and the code is being executed in RStudio. The
> RStudio version I am currently using is Version 2022.02.0 Build 443
> "Prairie Trillium" Release (9f796939, 2022-02-16) for Windows.
>
> My PC specs:
> Processor: Intel(R) Core(TM) i5-10310U CPU @ 1.70 GHz
> Installed RAM: 16.0 GB (15.6 GB usable)
> System type: 64-bit operating system, x64-based processor
> Local Disc(C:) Free Space: 274 GB
>
> I am wondering if there is/are a set of system variable(s) or something I
> could do to improve the performance of the program.
>
> It is really odd this code has taken this much (and it is still running).
>
> Any help and/or guidance would be greatly appreciated.
>
> Best regards,
> Paul
>
>
>
>
> #performing 1,000,000 simulations 10 times
> num_trials_6 = 100
> dice_rolls_6 = num_trials_6*12
> num_dice_6 = 1
> dice_sides_6 = 6
>
> prob_frame_6 <- data.frame(matrix(ncol = 10, nrow = 1))
>
> k <- 0
> while(k < 10){
> dice_simul_6 = data.frame(dice(rolls = dice_rolls_6, ndice = num_dice_6,
> sides = dice_sides_6, plot.it = FALSE))
>
> #constructing matrix containing results of all dice rolls by month
> prob_matrix_6 <- data.frame(matrix(dice_simul_6[,1], ncol = 12, byrow =
> TRUE))
>
> #naming each column by it's corresponding month name
> colnames(prob_matrix_6) <-
> c("Jan","Feb","Mar","Apr","May","Jun","Jul","Aug","Sep","Oct","Nov","Dec")
>
>
> #assigning each person´s name depending on the number showed in the dice
> once rolled
> for (i in 1:nrow(prob_matrix_6)){
> for (j in 1:ncol(prob_matrix_6)){
> if (prob_matrix_6[i,j] == 1){
> prob_matrix_6[i,j] = "Alice"
> }
> if (prob_matrix_6[i,j] == 2){
> prob_matrix_6[i,j] = "Bob"
> }
> if (prob_matrix_6[i,j] == 3){
> prob_matrix_6[i,j] = "Charlie"
> }
> if (prob_matrix_6[i,j] == 4){
> prob_matrix_6[i,j] = "Don"
> }
> if (prob_matrix_6[i,j] == 5){
> prob_matrix_6[i,j] = "Ellen"
> }
> if (prob_matrix_6[i,j] == 6){
> prob_matrix_6[i,j] = "Fred"
> }
>
> }
> }
>
> #calculating column which will have a 1 if trial was successful and a 0
> otherwise
> prob_matrix_6['success'] <- for (i in 1:nrow(prob_matrix_6)){
> if (("Alice" %in% prob_matrix_6[i,]) & ("Bob" %in% prob_matrix_6[i,]) &
> ("Charlie" %in% prob_matrix_6[i,]) & ("Don" %in% prob_matrix_6[i,]) &
> ("Ellen" %in% prob_matrix_6[i,]) & ("Fred" %in% prob_matrix_6[i,])){
> prob_matrix_6[i,13] = 1
> }else{
> prob_matrix_6[i,13] = 0
> }
> }
>
> #relabeling column v13 so that its new name is success
> colnames(prob_matrix_6)[13] <- "success"
>
>
> #calculating probability of success
>
> p6 = sum(prob_matrix_6$success)/nrow(prob_matrix_6)
> prob_frame_6 <- cbind(prob_frame_6, p6)
>
> k = k + 1
>
> }
>
> prob_frame_6 <- prob_frame_6[11:20]
> colnames(prob_frame_6) <-
> c("p1","p2","p3","p4","p5","p6","p7","p8","p9","p10")
> average_prob_frame_6 <- rowMeans(prob_frame_6)
> trial_100_10_frame <- cbind(prob_frame_6, average_prob_frame_6)
> final_frame_6 <- trial_100_10_frame
> colnames(final_frame_6) <-
> c("p1","p2","p3","p4","p5","p6","p7","p8","p9","p10", "avg_prob_frame_5")
>
> write.csv(final_frame_6, "OneMillion_Trials_Ten_Times_Results.csv")
> print(final_frame_6)
> print(paste("The average probability of success when doing 1,000,000 trials
> 10 times is:", average_prob_frame_6))
>
> [[alternative HTML version deleted]]
>
> __
> R-help@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.
__
R-help@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.
