v0.9.6 - Use RcppMLPACK directly instead of via a wrapper, emstreeR.

gagolews · Jan 4, 2021 · 346a88d · 346a88d
1 parent 87e8769
commit 346a88d
Show file tree

Hide file tree

Showing 60 changed files with 207 additions and 210 deletions.
diff --git a/.github/workflows/py.yml b/.github/workflows/py.yml
@@ -25,7 +25,7 @@ jobs:
         sudo apt-get update
         sudo apt-get -y upgrade
         sudo apt-get -y install libcurl4-openssl-dev python3-rpy2 libmlpack-dev r-base-dev python3-dev pandoc
-        sudo Rscript -e "install.packages(c('genie', 'Rcpp', 'tinytest', 'emstreeR', 'knitr', 'roxygen2'))"
+        sudo Rscript -e "install.packages(c('genie', 'Rcpp', 'tinytest', 'RcppArmadillo', 'RcppMLPACK', 'knitr', 'roxygen2'))"
         #sudo Rscript -e "devtools::install_github('gagolews/Rd2rst')"
         python -m pip install --upgrade pip
         if [ -f requirements.txt ]; then pip install -r requirements.txt --upgrade; fi

diff --git a/.github/workflows/r.yml b/.github/workflows/r.yml
@@ -16,7 +16,7 @@ jobs:
         sudo apt-get update
         sudo apt-get -y upgrade
         sudo apt-get -y install libcurl4-openssl-dev python3-rpy2 libmlpack-dev r-base-dev
-        sudo Rscript -e "install.packages(c('genie', 'Rcpp', 'tinytest', 'emstreeR', 'roxygen2'))"
+        sudo Rscript -e "install.packages(c('genie', 'Rcpp', 'tinytest', 'RcppArmadillo', 'RcppMLPACK', 'roxygen2'))"
     - name: Test and check R
       run: |
         sudo make r-test

diff --git a/DESCRIPTION b/DESCRIPTION
@@ -2,7 +2,7 @@ Package: genieclust
 Type: Package
 Title: The Genie++ Hierarchical Clustering Algorithm with Noise Points Detection
 Version: 0.9.6
-Date: 2021-01-04
+Date: 2021-01-05
 Authors@R: c(
     person("Marek", "Gagolewski",
         role = c("aut", "cre", "cph"),
@@ -37,9 +37,8 @@ Imports:
     stats,
     utils
 Suggests:
-    datasets,
-    emstreeR
-LinkingTo: Rcpp
+    datasets
+LinkingTo: Rcpp, RcppArmadillo, RcppMLPACK
 Encoding: UTF-8
 SystemRequirements: OpenMP, C++11
 RoxygenNote: 7.1.1
diff --git a/NEWS b/NEWS
@@ -1,11 +1,11 @@
 # `genieclust` Package NEWS
 
 
-## genieclust 0.9.6 (under development)
+## genieclust 0.9.6 (2021-01-05)
 
 -   [Python] Python >= 3.7 is now required (implied by `numpy`).
 
--   ... (nothing yet) ...
+-   [R] Use `RcppMLPACK` directly instead of via a wrapper, `emstreeR`.
 
 
 ## genieclust 0.9.5 (2021-01-04)

diff --git a/R/RcppExports.R b/R/RcppExports.R
@@ -149,6 +149,10 @@ pair_sets_index <- function(x, y = NULL) {
     .Call(`_genieclust_pair_sets_index`, x, y)
 }
 
+.emst_mlpack <- function(X) {
+    .Call(`_genieclust___emst_mlpack`, X)
+}
+
 .gclust <- function(mst, gini_threshold, verbose) {
     .Call(`_genieclust___gclust`, mst, gini_threshold, verbose)
 }

diff --git a/R/gclust.R b/R/gclust.R
@@ -2,7 +2,7 @@
 
 # ############################################################################ #
 #                                                                              #
-#   Copyleft (C) 2020, Marek Gagolewski <https://www.gagolewski.com>           #
+#   Copyleft (C) 2020-2021, Marek Gagolewski <https://www.gagolewski.com>      #
 #                                                                              #
 #                                                                              #
 #   This program is free software: you can redistribute it and/or modify       #
@@ -165,7 +165,7 @@
 #' pair_sets_index(y_test, y_pred)
 #'
 #' # Fast for low-dimensional Euclidean spaces:
-#' if (require("emstreeR")) h <- gclust(emst_mlpack(X))
+#' h <- gclust(emst_mlpack(X))
 #'
 #' @rdname gclust
 #' @export

diff --git a/R/mst.R b/R/mst.R
@@ -2,7 +2,7 @@
 
 # ############################################################################ #
 #                                                                              #
-#   Copyleft (C) 2020, Marek Gagolewski <https://www.gagolewski.com>           #
+#   Copyleft (C) 2020-2021, Marek Gagolewski <https://www.gagolewski.com>      #
 #                                                                              #
 #                                                                              #
 #   This program is free software: you can redistribute it and/or modify       #
@@ -177,18 +177,12 @@ registerS3method("mst", "dist",    "mst.dist")
 #'
 #' @description
 #' Provides access to an implementation of the Dual-Tree Borůvka
-#' algorithm based on kd-trees. It is fast for (very) low-dimensional
+#' algorithm based on kd-trees from MLPACK. It is fast for (very) low-dimensional
 #' Euclidean spaces. For higher dimensional spaces (say, over 5 features)
-#' or other metrics,
-#' use the parallelised Prim-like algorithm implemented in \code{\link{mst}()}.
+#' or other metrics, use the parallelised Prim-like algorithm implemented
+#' in \code{\link{mst}()}.
 #'
 #'
-#' @details
-#' Calls \code{emstreeR::mlpack_mst()} and converts the result
-#' so that it is compatible with the output of \code{\link{mst}()}.
-#'
-#' If the \code{emstreeR} package is not available, an error is generated.
-#'
 #' @param X a numeric matrix (or an object coercible to one,
 #'     e.g., a data frame with numeric-like columns)
 #' @param verbose logical; whether to print diagnostic messages
@@ -209,20 +203,16 @@ registerS3method("mst", "dist",    "mst.dist")
 emst_mlpack <- function(X, verbose=FALSE)
 {
     X <- as.matrix(X)
-    if (requireNamespace("emstreeR", quietly=TRUE)) {
-
-        if (!verbose)
-            capture.output({mst <- emstreeR:::mlpack_mst(X)})
-        else
-            mst <- emstreeR:::mlpack_mst(X)
-
-        structure(t(mst),
-            class="mst",
-            method="euclidean",
-            Labels=dimnames(X)[[1]]
-        )
-
-    } else {
-        stop("emstreeR` package is not installed.")
-    }
+
+    if (!verbose)
+        capture.output({mst <- .emst_mlpack(X)})
+    else
+        mst <- .emst_mlpack(X)
+
+    structure(
+        mst,
+        class="mst",
+        method="euclidean",
+        Labels=dimnames(X)[[1]]
+    )
 }
diff --git a/devel/sphinx/conf.py b/devel/sphinx/conf.py
@@ -25,7 +25,7 @@
 # -- Project information -----------------------------------------------------
 
 project = 'genieclust'
-copyright = '2020, Marek Gagolewski'
+copyright = '2021, Marek Gagolewski'
 author = 'Marek Gagolewski'
 
 # The full version, including alpha/beta/rc tags

diff --git a/devel/sphinx/rapi/emst_mlpack.rst b/devel/sphinx/rapi/emst_mlpack.rst
@@ -4,7 +4,7 @@ emst_mlpack: Euclidean Minimum Spanning Tree
 Description
 ~~~~~~~~~~~
 
-Provides access to an implementation of the Dual-Tree Borůvka algorithm based on kd-trees. It is fast for (very) low-dimensional Euclidean spaces. For higher dimensional spaces (say, over 5 features) or other metrics, use the parallelised Prim-like algorithm implemented in `mst() <mst.html>`__.
+Provides access to an implementation of the Dual-Tree Borůvka algorithm based on kd-trees from MLPACK. It is fast for (very) low-dimensional Euclidean spaces. For higher dimensional spaces (say, over 5 features) or other metrics, use the parallelised Prim-like algorithm implemented in `mst() <mst.html>`__.
 
 Usage
 ~~~~~
@@ -22,13 +22,6 @@ Arguments
 | ``verbose`` | logical; whether to print diagnostic messages                                                  |
 +-------------+------------------------------------------------------------------------------------------------+
 
-Details
-~~~~~~~
-
-Calls ``emstreeR::mlpack_mst()`` and converts the result so that it is compatible with the output of `mst() <mst.html>`__.
-
-If the ``emstreeR`` package is not available, an error is generated.
-
 Value
 ~~~~~
 

diff --git a/devel/sphinx/rapi/gclust.rst b/devel/sphinx/rapi/gclust.rst
@@ -146,4 +146,4 @@ Examples
    pair_sets_index(y_test, y_pred)
 
    # Fast for low-dimensional Euclidean spaces:
-   if (require("emstreeR")) h <- gclust(emst_mlpack(X))
+   h <- gclust(emst_mlpack(X))
diff --git a/devel/tinytest/test-gclust.R b/devel/tinytest/test-gclust.R
@@ -19,10 +19,8 @@ if (require("genie")) {
     #for (k in 2:20) expect_equal(adjusted_rand_score(cutree(h1, k), cutree(h2, k)), 1.0)
 }
 
-if (suppressMessages(require("emstreeR"))) {
-    print(system.time(h2 <- gclust(emst_mlpack(X, verbose=verbose), verbose=verbose)))
-    #for (k in 2:20) expect_equal(adjusted_rand_score(cutree(h1, k), cutree(h2, k)), 1.0)
-}
+print(system.time(h2 <- gclust(emst_mlpack(X, verbose=verbose), verbose=verbose)))
+#for (k in 2:20) expect_equal(adjusted_rand_score(cutree(h1, k), cutree(h2, k)), 1.0)
 
 
 # 2020-06-21

diff --git a/devel/tinytest/test-mst.R b/devel/tinytest/test-mst.R
@@ -23,14 +23,11 @@ expect_true(all(t2[,1] < t2[,2]))
 expect_true(all(diff(t2[,3])>=0))
 
 
-
-if (require("emstreeR")) {
-    print(system.time(t2 <- emst_mlpack(X)))
-    expect_equal(t1[,1], t2[,1])
-    expect_equal(t1[,2], t2[,2])
+print(system.time(t2 <- emst_mlpack(X)))
+expect_equal(t1[,1], t2[,1])
+expect_equal(t1[,2], t2[,2])
 #     print(abs(sum(t1[,3])-sum(t2[,3])))
-    expect_true(abs(sum(t1[,3])-sum(t2[,3]))<1e-12)
-}
+expect_true(abs(sum(t1[,3])-sum(t2[,3]))<1e-12)
 
 
 
@@ -50,7 +47,7 @@ for (M in c(1, 5, 10)) {
 #         print(abs(sum(t1[,3])-sum(t2[,3])))
         expect_true(abs(sum(t1[,3])-sum(t2[,3]))<1e-12)
 
-        if (distance == "euclidean" && M == 1 && require("emstreeR")) {
+        if (distance == "euclidean" && M == 1) {
             t2 <- emst_mlpack(X)
             expect_equal(t1[,1], t2[,1])
             expect_equal(t1[,2], t2[,2])

diff --git a/docs/_sources/rapi/emst_mlpack.rst.txt b/docs/_sources/rapi/emst_mlpack.rst.txt
@@ -4,7 +4,7 @@ emst_mlpack: Euclidean Minimum Spanning Tree
 Description
 ~~~~~~~~~~~
 
-Provides access to an implementation of the Dual-Tree Borůvka algorithm based on kd-trees. It is fast for (very) low-dimensional Euclidean spaces. For higher dimensional spaces (say, over 5 features) or other metrics, use the parallelised Prim-like algorithm implemented in `mst() <mst.html>`__.
+Provides access to an implementation of the Dual-Tree Borůvka algorithm based on kd-trees from MLPACK. It is fast for (very) low-dimensional Euclidean spaces. For higher dimensional spaces (say, over 5 features) or other metrics, use the parallelised Prim-like algorithm implemented in `mst() <mst.html>`__.
 
 Usage
 ~~~~~
@@ -22,13 +22,6 @@ Arguments
 | ``verbose`` | logical; whether to print diagnostic messages                                                  |
 +-------------+------------------------------------------------------------------------------------------------+
 
-Details
-~~~~~~~
-
-Calls ``emstreeR::mlpack_mst()`` and converts the result so that it is compatible with the output of `mst() <mst.html>`__.
-
-If the ``emstreeR`` package is not available, an error is generated.
-
 Value
 ~~~~~
 

diff --git a/docs/_sources/rapi/gclust.rst.txt b/docs/_sources/rapi/gclust.rst.txt
@@ -146,4 +146,4 @@ Examples
    pair_sets_index(y_test, y_pred)
 
    # Fast for low-dimensional Euclidean spaces:
-   if (require("emstreeR")) h <- gclust(emst_mlpack(X))
+   h <- gclust(emst_mlpack(X))
diff --git a/docs/_static/documentation_options.js b/docs/_static/documentation_options.js
@@ -1,6 +1,6 @@
 var DOCUMENTATION_OPTIONS = {
     URL_ROOT: document.getElementById("documentation_options").getAttribute('data-url_root'),
-    VERSION: '0.9.6.dev0',
+    VERSION: '0.9.6',
     LANGUAGE: 'None',
     COLLAPSE_INDEX: false,
     BUILDER: 'html',

diff --git a/docs/genieclust.html b/docs/genieclust.html
@@ -7,7 +7,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>Python Package genieclust Reference &mdash; genieclust 0.9.6.dev0 documentation</title>
+  <title>Python Package genieclust Reference &mdash; genieclust 0.9.6 documentation</title>
 
 
 
@@ -66,7 +66,7 @@
 
 
               <div class="version">
-                0.9.6.dev0
+                0.9.6
               </div>
 
 
@@ -261,7 +261,7 @@ <h1>Python Package <cite>genieclust</cite> Reference<a class="headerlink" href="
   <div role="contentinfo">
     <p>
 
-        &copy; Copyright 2020, Marek Gagolewski
+        &copy; Copyright 2021, Marek Gagolewski
 
     </p>
   </div>

diff --git a/docs/genieclust_compare_partitions.html b/docs/genieclust_compare_partitions.html
@@ -7,7 +7,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>genieclust.compare_partitions &mdash; genieclust 0.9.6.dev0 documentation</title>
+  <title>genieclust.compare_partitions &mdash; genieclust 0.9.6 documentation</title>
 
 
 
@@ -66,7 +66,7 @@
 
 
               <div class="version">
-                0.9.6.dev0
+                0.9.6
               </div>
 
 
@@ -822,7 +822,7 @@
   <div role="contentinfo">
     <p>
 
-        &copy; Copyright 2020, Marek Gagolewski
+        &copy; Copyright 2021, Marek Gagolewski
 
     </p>
   </div>

diff --git a/docs/genieclust_genie.html b/docs/genieclust_genie.html
@@ -7,7 +7,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>genieclust.Genie &mdash; genieclust 0.9.6.dev0 documentation</title>
+  <title>genieclust.Genie &mdash; genieclust 0.9.6 documentation</title>
 
 
 
@@ -66,7 +66,7 @@
 
 
               <div class="version">
-                0.9.6.dev0
+                0.9.6
               </div>
 
 
@@ -610,7 +610,7 @@ <h1>genieclust.Genie<a class="headerlink" href="#genieclust-genie" title="Permal
   <div role="contentinfo">
     <p>
 
-        &copy; Copyright 2020, Marek Gagolewski
+        &copy; Copyright 2021, Marek Gagolewski
 
     </p>
   </div>

diff --git a/docs/genieclust_inequity.html b/docs/genieclust_inequity.html
@@ -7,7 +7,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>genieclust.inequity &mdash; genieclust 0.9.6.dev0 documentation</title>
+  <title>genieclust.inequity &mdash; genieclust 0.9.6 documentation</title>
 
 
 
@@ -66,7 +66,7 @@
 
 
               <div class="version">
-                0.9.6.dev0
+                0.9.6
               </div>
 
 
@@ -379,7 +379,7 @@
   <div role="contentinfo">
     <p>
 
-        &copy; Copyright 2020, Marek Gagolewski
+        &copy; Copyright 2021, Marek Gagolewski
 
     </p>
   </div>

diff --git a/docs/genieclust_internal.html b/docs/genieclust_internal.html
@@ -7,7 +7,7 @@
 
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
 
-  <title>genieclust.internal &mdash; genieclust 0.9.6.dev0 documentation</title>
+  <title>genieclust.internal &mdash; genieclust 0.9.6 documentation</title>
 
 
 
@@ -66,7 +66,7 @@
 
 
               <div class="version">
-                0.9.6.dev0
+                0.9.6
               </div>
 
 
@@ -540,7 +540,7 @@
   <div role="contentinfo">
     <p>
 
-        &copy; Copyright 2020, Marek Gagolewski
+        &copy; Copyright 2021, Marek Gagolewski
 
     </p>
   </div>

diff --git a/docs/genieclust_plots-1.pdf b/docs/genieclust_plots-1.pdf
diff --git a/docs/genieclust_plots-2.pdf b/docs/genieclust_plots-2.pdf
diff --git a/docs/genieclust_plots-3.pdf b/docs/genieclust_plots-3.pdf
diff --git a/docs/genieclust_plots-4.pdf b/docs/genieclust_plots-4.pdf