Merge pull request #14 from tanaylab/feat@parallel

Added parallelism
Changed the default to id_column = FALSE

.Rbuildignore

@@ -16,3 +16,5 @@
 ^cran-comments\.md$
 ^CRAN-SUBMISSION$
 ^\.github$
+^work$
+^\.vscode$

.gitignore

@@ -8,3 +8,5 @@ README.html
 /doc/
 /Meta/
 src/symbols.rds
+work
+.vscode

DESCRIPTION

@@ -1,36 +1,42 @@
 Package: tglkmeans
 Title: Efficient Implementation of K-Means++ Algorithm
-Version: 0.3.12
-Authors@R: 
-    c(person(given = "Aviezer",
-             family = "Lifshitz",
-             role = c("aut", "cre"),
-             email = "[email protected]"),
-      person(given = "Amos",
-             family = "Tanay",
-             role = "aut"),             
-      person("Weizmann Institute of Science", role="cph"))
-Author: Aviezer Lifshitz [aut, cre], Amos Tanay [aut], Weizmann Institute of Science [cph]
+Version: 0.4.0
+Authors@R: c(
+    person("Aviezer", "Lifshitz", , "[email protected]", role = c("aut", "cre")),
+    person("Amos", "Tanay", role = "aut"),
+    person("Weizmann Institute of Science", role = "cph")
+  )
+Author: Aviezer Lifshitz [aut, cre], Amos Tanay [aut], Weizmann Institute
+    of Science [cph]
 Maintainer: Aviezer Lifshitz <[email protected]>
-Description: Efficient implementation of K-Means++ algorithm. For more information see (1) "kmeans++ the advantages of the k-means++ algorithm" by David Arthur and Sergei Vassilvitskii (2007), Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms, Society for Industrial and Applied Mathematics, Philadelphia, PA, USA, pp. 1027-1035, and (2) "The Effectiveness of Lloyd-Type Methods for the k-Means Problem" by Rafail Ostrovsky, Yuval Rabani, Leonard J. Schulman and Chaitanya Swamy <doi:10.1145/2395116.2395117>.
+Description: Efficient implementation of K-Means++ algorithm. For more
+    information see (1) "kmeans++ the advantages of the k-means++
+    algorithm" by David Arthur and Sergei Vassilvitskii (2007),
+    Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete
+    algorithms, Society for Industrial and Applied Mathematics,
+    Philadelphia, PA, USA, pp. 1027-1035, and (2) "The Effectiveness of
+    Lloyd-Type Methods for the k-Means Problem" by Rafail Ostrovsky, Yuval
+    Rabani, Leonard J. Schulman and Chaitanya Swamy
+    <doi:10.1145/2395116.2395117>.
 License: MIT + file LICENSE
 BugReports: https://github.com/tanaylab/tglkmeans/issues
 URL: https://tanaylab.github.io/tglkmeans/, https://github.com/tanaylab/tglkmeans
-OS_type: unix
 Depends: 
     R (>= 4.0.0)
 Imports: 
-    Rcpp (>= 0.12.11),
+    cli,
     doFuture,
+    dplyr (>= 0.5.0),
     future,
-    dplyr (>= 0.5.0),    
-    ggplot2 (>= 2.2.0),    
+    ggplot2 (>= 2.2.0),
     magrittr,
-    tibble (>= 3.2.1),
-    parallel (>= 3.3.2),    
+    parallel (>= 3.3.2),
     plyr (>= 1.8.4),
-    purrr (>= 0.2.0),    
-    tgstat (>= 1.0.0)
+    purrr (>= 0.2.0),
+    Rcpp (>= 0.12.11),
+    RcppParallel,
+    tgstat (>= 1.0.0),
+    tibble (>= 3.2.1)
 Suggests: 
     covr,
     knitr,
@@ -39,10 +45,14 @@ Suggests:
     testthat,
     withr
 LinkingTo: 
-    Rcpp
+    Rcpp,
+    RcppParallel
 VignetteBuilder: 
     knitr
+Config/testthat/edition: 3
+Config/testthat/parallel: false
 Encoding: UTF-8
 NeedsCompilation: yes
-Packaged: 2018-02-25 13:32:32 UTC; aviezerl
+OS_type: unix
 RoxygenNote: 7.2.3
+SystemRequirements: GNU make

NAMESPACE

@@ -8,6 +8,12 @@ export(tglkmeans.set_parallel)
 import(dplyr)
 import(ggplot2)
 importFrom(Rcpp,sourceCpp)
+importFrom(RcppParallel,RcppParallelLibs)
+importFrom(cli,cli_abort)
+importFrom(cli,cli_alert)
+importFrom(cli,cli_alert_info)
+importFrom(cli,cli_alert_success)
+importFrom(cli,cli_warn)
 importFrom(grDevices,colorRampPalette)
 importFrom(grDevices,dev.off)
 importFrom(magrittr,"%>%")

NEWS.md

@@ -1,16 +1,22 @@
+# tglkmeans 0.4.0
+
+* Default of `id_column` parameter was changed to `FALSE`. Note that this is a breaking change, and if you want to use an id column, you need to set it explicitly to `TRUE`.
+* Use R random number generator instead of C++11 random number generator. For backwards compatibility, the old random number generator can be used by setting `use_cpp_random` to `TRUE`.
+* Added parallelization using `RcppParallel`. 
+
 # tglkmeans 0.3.12
 
 * Added validity checks for `k` and the number of observations. 
 
-# tgkmeans 0.3.11
+# tglkmeans 0.3.11
 
 * Changed pkgdoc, see: https://github.com/r-lib/roxygen2/issues/1491.
 
-# tgkmeans 0.3.10
+# tglkmeans 0.3.10
 
 * Removed broken link to one of the references in the description.
 
-# tgkmeans 0.3.9
+# tglkmeans 0.3.9
 
 * Remove empty clusters. This may happen when the number of clusters is larger than the number of observations, and currently caused an error in the reordering step.
 

R/RcppExports.R

@@ -9,7 +9,7 @@ reduce_num_trials <- function(boot_nodes_l, cc_mat) {
     invisible(.Call('_tglkmeans_reduce_num_trials', PACKAGE = 'tglkmeans', boot_nodes_l, cc_mat))
 }
 
-TGL_kmeans_cpp <- function(ids, mat, k, metric, max_iter = 40, min_delta = 0.0001, random_seed = TRUE, seed = -1L) {
-    .Call('_tglkmeans_TGL_kmeans_cpp', PACKAGE = 'tglkmeans', ids, mat, k, metric, max_iter, min_delta, random_seed, seed)
+TGL_kmeans_cpp <- function(ids, mat, k, metric, max_iter = 40, min_delta = 0.0001, use_cpp_random = FALSE, seed = -1L) {
+    .Call('_tglkmeans_TGL_kmeans_cpp', PACKAGE = 'tglkmeans', ids, mat, k, metric, max_iter, min_delta, use_cpp_random, seed)
 }