Last active
March 4, 2016 19:17
-
-
Save kylebaron/4b5b7d970b6f042d07c9 to your computer and use it in GitHub Desktop.
Set RNGkind in worker
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
library(shiny) | |
library(mrgsolve) | |
library(dplyr) | |
library(magrittr) | |
library(parallel) | |
RNGkind("L'Ecuyer-CMRG") | |
set.seed(101) | |
mc.reset.stream() | |
##' A model to simulate ETAs | |
code <- ' | |
$OMEGA 0 0 | |
$TABLE | |
table(ETA1) = ETA(1); | |
table(ETA2) = ETA(2); | |
' | |
mod <- mread("shinymc",tempdir(),code,warn=FALSE) | |
##' UI ########################################### | |
ui<- fluidPage( | |
titlePanel("Shiny MC"), | |
sidebarLayout( | |
sidebarPanel( | |
sliderInput("OM1", "OMEGA 1",0,4,1,0.25), | |
sliderInput("OM2", "OMEGA 2",0,4,1,.25), | |
sliderInput("n", "N rep", 1,100,10,1), | |
sliderInput("N", "N subj", 1, 100,10,1), | |
sliderInput("mccores", "mc.cores", 1,4,2,1), | |
sliderInput("seed", "set.seed",1,50,25,1) | |
), | |
mainPanel(tableOutput("table") | |
) | |
) | |
) | |
##' Simulate | |
##' The KEY is to set the RNGkind on the worker | |
sim <- function(i,x,idata) { | |
RNGkind("L'Ecuyer-CMRG") | |
x %>% idata_set(idata) %>% mrgsim(end=-1) %>% mutate(irep=i) | |
} | |
##' Summarize simulations | |
smry <- function(x) { | |
x %>% summarise(nETA1 = n_distinct(ETA1), | |
vETA1 = var(ETA1), | |
sETA1 = sum(ETA1), | |
nETA2 = n_distinct(ETA2), | |
vETA2 = var(ETA2), | |
sETA2 = sum(ETA2), | |
n=n()) | |
} | |
##' SERVER ########################################### | |
server<-function(input, output) { | |
##' For now, be safe | |
RNGkind("L'Ecuyer-CMRG") | |
output$table <- renderTable({ | |
idata <- data_frame(ID=1:input$N) | |
mod %<>% omat(dmat(input$OM1,input$OM2)) | |
##' For now, be safe | |
RNGkind("L'Ecuyer-CMRG") | |
set.seed(as.integer(input$seed)) | |
mc.reset.stream() | |
mclapply(1:input$n, | |
mc.cores=input$mccores, | |
sim, | |
mod, | |
idata) %>% bind_rows %>% smry | |
}) | |
} | |
##' Run the shiny app | |
shinyApp(ui = ui, server = server) | |
##' Run the shiny app simulation outside of shiny | |
RNGkind("L'Ecuyer-CMRG") | |
set.seed(101) | |
mc.reset.stream() | |
out <- mclapply(1:10,mc.cores=4,sim, | |
mod %>% omat(dmat(1,1)), | |
data_frame(ID=1:100),FALSE) %>% bind_rows | |
out %>% smry | |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment