To test the performance of calculation applying on dense dfm matrix versus on sparse dfm matrix versus using multi-thread (on sparse dfm)
require(quanteda, quietly = TRUE, warn.conflicts = FALSE) library("quanteda.textmodels") ie2010dfm <- dfm(data_corpus_irishbudget2010, verbose = FALSE) wfm_sparse <- function(x){ textmodel_wordfish(x, dir = c(6,5), threads = 1) } wfm_sparse_mt <- function(x){ threads <- RcppParallel::defaultNumThreads() quanteda_options(threads = threads - 1) textmodel_wordfish(x, dir = c(6,5)) } microbenchmark::microbenchmark( wfm_s_mt = wfm_sparse_mt(ie2010dfm), wfm_s = wfm_sparse(ie2010dfm), wfm_d = textmodel_wordfish(ie2010dfm, dir = c(6,5), sparse = FALSE), times=20, unit = 'relative')
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.