{"id":37,"date":"2013-05-29T13:48:04","date_gmt":"2013-05-29T11:48:04","guid":{"rendered":"http:\/\/candrea.ch\/blog\/?p=37"},"modified":"2013-05-29T14:23:09","modified_gmt":"2013-05-29T12:23:09","slug":"parallel-computing-unidimtest-in-irt","status":"publish","type":"post","link":"https:\/\/candrea.ch\/blog\/parallel-computing-unidimtest-in-irt\/","title":{"rendered":"Parallel computing unidimTest in IRT"},"content":{"rendered":"

In the R<\/a> package ltm<\/a>\u00a0by\u00a0Dimitris Rizopoulos there is a function called unidimTest()<\/a>. Computations of this function are very power consuming due the\u00a0Monte Carlo procedure used inside. Without parallelizing only one core of your (surely) multicore computer is used for this computation. A simple modification of unidimTest()<\/a> makes it possible to use the quite easy to use R<\/a>\u00a0package foreach<\/a> for parallel computing using the R<\/a>\u00a0packages parallel<\/a> and doParallel<\/a>. See also parallel-r-loops-for-windows-and-linux<\/a>\u00a0by Vik Paruchuri.<\/p>\n

I have changed the lines\u00a0of\u00a0unidimTest()<\/a><\/p>\n

for (b in 1:B) {\r\n  if (!missing(object)) \r\n    z.vals <- rnorm(n, ablts, se.ablts)\r\n  data.new <- rmvlogis(n, parms, IRT = IRT, z.vals = z.vals)\r\n  T.boot[b, ] <- eigenRho(data.new)$ev\r\n}<\/pre>\n
to<\/p>\n
T.boot[,] <- foreach(b=1:B, .combine=\"rbind\", .errorhandling=\"pass\") %dopar%\r\n{\r\n  #if (!missing(object)) \r\n  z.vals <- rnorm(n, ablts, se.ablts)\r\n  data.new <- rmvlogis(n, parms, IRT = IRT, z.vals = z.vals)\r\n  eigenRho(data.new)$ev   \t\r\n}<\/pre>\n
and named the function unidimTest.p().<\/p>\n
The function is called as<\/p>\n
library(foreach)\r\nlibrary(parallel)\r\nlibrary(doParallel)\r\n\r\nregisterDoParallel(cores=4) # put in here how many cores you want to be used\r\n\r\nset.seed(666)\r\nproc <- proc.time()\r\nuni.p <- unidimTest.p(fit.rasch.u, B=100)\r\nproc.time() - proc<\/pre>\n
and works at least on an Mac OS X (10.8.3) environment with\u00a0R<\/a>\u00a0version 3.0.1.<\/p>\n
 <\/p>\n
 <\/p>\n","protected":false},"excerpt":{"rendered":"
In the R package ltm\u00a0by\u00a0Dimitris Rizopoulos there is a function called unidimTest(). Computations of this function are very power consuming due the\u00a0Monte Carlo procedure used inside. Without parallelizing only one core of your (surely) multicore computer is used for this computation. A simple modification of unidimTest() makes it possible to use the quite easy to … Continue reading Parallel computing unidimTest in IRT<\/span> →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3],"tags":[],"class_list":["post-37","post","type-post","status-publish","format-standard","hentry","category-r"],"_links":{"self":[{"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/posts\/37"}],"collection":[{"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/comments?post=37"}],"version-history":[{"count":8,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/posts\/37\/revisions"}],"predecessor-version":[{"id":45,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/posts\/37\/revisions\/45"}],"wp:attachment":[{"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/media?parent=37"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/categories?post=37"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/candrea.ch\/blog\/wp-json\/wp\/v2\/tags?post=37"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}