Merge pull request #57 from imbs-hl/win_threads

Windows multithreading support
imbs-hl · May 2, 2016 · 6b20ddd · 6b20ddd
2 parents d980670 + c5b597c
commit 6b20ddd
Show file tree

Hide file tree

Showing 11 changed files with 43 additions and 26 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -1,3 +1,6 @@
+##### Version 0.4.2
+* Add Windows multithreading support for new toolchain
+
 ##### Version 0.4.1
 * Runtime improvement for regression forests on classification data
 

diff --git a/README.md b/README.md
@@ -17,7 +17,7 @@ To install the Ranger R package from CRAN, just run
 install.packages("ranger”)
 ```
 
-R version >= 3.1 is required. Note that, for now, no multithreading is supported in the R version on Windows platforms (the compiler in RTools is too old).
+R version >= 3.1 is required. Note that, for now, R-devel and the new RTools toolchain is required for multithreading on Windows platforms (or install a binary version).
 
 To install the C++ version of Ranger in Linux or Mac OS X you will need a compiler supporting C++11 (i.e. gcc >= 4.7 or Clang >= 3.0) and Cmake. To build start a terminal from the Ranger main directory and run the following commands
 
@@ -55,6 +55,7 @@ ranger --verbose --file data.dat --depvarname Species --treetype 1 --ntree 1000
 If you find any bugs, or if you experience any crashes, please report to us. If you have any questions just ask, we won't bite. 
 
 ### References
+* Wright, M. N. & Ziegler, A. (2016). ranger: A Fast Implementation of Random Forests for High Dimensional Data in C++ and R. Journal of Statistical Software, in press. http://arxiv.org/abs/1508.04409.
 * Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
 * Ishwaran, H., Kogalur, U. B., Blackstone, E. H., & Lauer, M. S. (2008). Random survival forests. The Annals of Applied Statistics, 841-860.
 * Malley, J. D., Kruppa, J., Dasgupta, A., Malley, K. G., & Ziegler, A. (2012). Probability machines: consistent probability estimation using nonparametric learning machines. Methods Inf Med, 51(1), 74.
diff --git a/ranger-r-package/ranger/DESCRIPTION b/ranger-r-package/ranger/DESCRIPTION
@@ -1,8 +1,8 @@
 Package: ranger
 Type: Package
 Title: A Fast Implementation of Random Forests
-Version: 0.4.1
-Date: 2016-04-28
+Version: 0.4.2
+Date: 2016-05-02
 Author: Marvin N. Wright
 Maintainer: Marvin N. Wright <wright@imbs.uni-luebeck.de>
 Description: A fast implementation of Random Forests, particularly suited for high dimensional data. Ensembles

diff --git a/ranger-r-package/ranger/NEWS b/ranger-r-package/ranger/NEWS
@@ -1,3 +1,6 @@
+##### Version 0.4.2
+* Add Windows multithreading support for new toolchain
+
 ##### Version 0.4.1
 * Runtime improvement for regression forests on classification data
 

diff --git a/ranger-r-package/ranger/R/ranger.R b/ranger-r-package/ranger/R/ranger.R
@@ -58,11 +58,12 @@
 ##' To use only the SNPs without sex or other covariates from the phenotype file, use \code{0} on the right hand side of the formula. 
 ##' Note that missing values are treated as an extra category while splitting.
 ##' 
-##' See \url{https://github.com/mnwright/ranger} for the development version.
+##' See \url{https://github.com/imbs-hl/ranger} for the development version.
 ##' 
-##' Notes:
+##' To use multithreading on Microsoft Windows platforms, there are currently two options:
 ##' \itemize{
-##'  \item Multithreading is currently not supported for Microsoft Windows platforms.
+##'  \item Use R-devel with the new toolchain
+##'  \item Install a binary version of ranger, download: \url{https://github.com/imbs-hl/ranger/releases}
 ##' }
 ##' 
 ##' @title Ranger

diff --git a/ranger-r-package/ranger/man/ranger.Rd b/ranger-r-package/ranger/man/ranger.Rd
diff --git a/ranger-r-package/ranger/src/AAA_check_cpp11.cpp b/ranger-r-package/ranger/src/AAA_check_cpp11.cpp
@@ -1,7 +1,6 @@
-
 #ifndef WIN_R_BUILD
 #if __cplusplus < 201103L
-#error Error: ranger requires a real C++11 compiler. You probably have to update gcc. 
+#error Error: ranger requires a real C++11 compiler. You probably have to update gcc.
 #endif
 #endif
 
diff --git a/source/src/Forest/Forest.cpp b/source/src/Forest/Forest.cpp
@@ -32,7 +32,7 @@
 #include <string>
 #include <ctime>
 #include <math.h>
-#ifndef WIN_R_BUILD
+#ifndef OLD_WIN_R_BUILD
 #include <thread>
 #include <chrono>
 #endif
@@ -191,7 +191,7 @@ void Forest::init(std::string dependent_variable_name, MemoryMode memory_mode, D
 
   // Set number of threads
   if (num_threads == DEFAULT_NUM_THREADS) {
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
     this->num_threads = 1;
 #else
     this->num_threads = std::thread::hardware_concurrency();
@@ -413,7 +413,7 @@ void Forest::grow() {
   variable_importance.resize(num_independent_variables, 0);
 
   // Grow trees in multiple threads
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
   progress = 0;
   clock_t start_time = clock();
   clock_t lap_time = clock();
@@ -476,7 +476,7 @@ void Forest::grow() {
 void Forest::predict() {
 
   // Predict trees in multiple threads and join the threads with the main thread
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
   progress = 0;
   clock_t start_time = clock();
   clock_t lap_time = clock();
@@ -516,7 +516,7 @@ void Forest::predict() {
 void Forest::computePredictionError() {
 
   // Predict trees in multiple threads
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
   progress = 0;
   clock_t start_time = clock();
   clock_t lap_time = clock();
@@ -550,7 +550,7 @@ void Forest::computePredictionError() {
 void Forest::computePermutationImportance() {
 
   // Compute tree permutation importance in multiple threads
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
   progress = 0;
   clock_t start_time = clock();
   clock_t lap_time = clock();
@@ -637,7 +637,7 @@ void Forest::computePermutationImportance() {
   }
 }
 
-#ifndef WIN_R_BUILD
+#ifndef OLD_WIN_R_BUILD
 void Forest::growTreesInThread(uint thread_idx, std::vector<double>* variable_importance) {
   if (thread_ranges.size() > thread_idx + 1) {
     for (size_t i = thread_ranges[thread_idx]; i < thread_ranges[thread_idx + 1]; ++i) {
@@ -813,7 +813,7 @@ void Forest::setAlwaysSplitVariables(std::vector<std::string>& always_split_vari
   }
 }
 
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
 void Forest::showProgress(std::string operation, clock_t start_time, clock_t& lap_time) {
 
   // Check for user interrupt

diff --git a/source/src/Forest/Forest.h b/source/src/Forest/Forest.h
@@ -33,7 +33,7 @@
 #include <iostream>
 #include <random>
 #include <ctime>
-#ifndef WIN_R_BUILD
+#ifndef OLD_WIN_R_BUILD
 #include <thread>
 #include <chrono>
 #include <mutex>
@@ -172,7 +172,7 @@ class Forest {
   void setAlwaysSplitVariables(std::vector<std::string>& always_split_variable_names);
 
   // Show progress every few seconds
-#ifdef WIN_R_BUILD
+#ifdef OLD_WIN_R_BUILD
   void showProgress(std::string operation, clock_t start_time, clock_t& lap_time);
 #else
   void showProgress(std::string operation);
@@ -207,7 +207,7 @@ class Forest {
   // Multithreading
   uint num_threads;
   std::vector<uint> thread_ranges;
-#ifndef WIN_R_BUILD
+#ifndef OLD_WIN_R_BUILD
   std::mutex mutex;
   std::condition_variable condition_variable;
 #endif

diff --git a/source/src/globals.h b/source/src/globals.h
@@ -1,6 +1,6 @@
 /*-------------------------------------------------------------------------------
 This file is part of Ranger.
-    
+
 Ranger is free software: you can redistribute it and/or modify
 it under the terms of the GNU General Public License as published by
 the Free Software Foundation, either version 3 of the License, or
@@ -14,13 +14,13 @@ GNU General Public License for more details.
 You should have received a copy of the GNU General Public License
 along with Ranger. If not, see <http://www.gnu.org/licenses/>.
 
-Written by: 
+Written by:
 
 Marvin N. Wright
 Institut für Medizinische Biometrie und Statistik
 Universität zu Lübeck
 Ratzeburger Allee 160
-23562 Lübeck 
+23562 Lübeck
 
 http://www.imbs-luebeck.de
 wright@imbs.uni-luebeck.de
@@ -33,6 +33,15 @@ wright@imbs.uni-luebeck.de
     TypeName(const TypeName&);             \
     void operator=(const TypeName&)
 
+// Old/new Win build
+#ifdef WIN_R_BUILD
+  #if __cplusplus < 201103L
+    #define OLD_WIN_R_BUILD
+  #else
+    #define NEW_WIN_R_BUILD
+  #endif
+#endif
+
 typedef unsigned int uint;
 
 // Tree types, probability is not selected by ID

diff --git a/source/src/version.h b/source/src/version.h
@@ -1,3 +1,3 @@
 #ifndef RANGER_VERSION
-#define RANGER_VERSION "0.4.1"
+#define RANGER_VERSION "0.4.2"
 #endif