Repository files navigation

Thewsrf is a parallelimplementation of the Weighted Subspace Random Forest algorithm (wsrf)ofXu et al. A novelvariable weighting method is used for variable subspace selection inplace of the traditional approach of random variable sampling. Thisnew approach is particularly useful in building models for highdimensional data---often consisting of thousands of variables.Parallel computation is used to take advantage of multi-core machinesand clusters of machines to build random forest models from highdimensional data with reduced elapsed times.

The package ships with ahtmlvignetteincluding more details and a few examples.

Currently, wsrf requiresR (>= 3.3.0),Rcpp (>= 0.10.2). For theuse of multi-threading, a C++ compiler withC++11 standard support ofthreads (for example,GCC 4.8.1) is required.Since the latest version of R has added support for C++11 on alloperating systems, we do not provide support for the old version of Rand C++ compiler without C++11 support. To install the latest versionof the package, from within R run:

R> install.packages("wsrf")

Previous version of wsrf provide support on systems without C++11 orusing Boost for multithreading. Though we do not provide support forthese options anymore, but still leave the usage here for someone withneeds of previous version of wsrf. The choice is available atinstallation time depending on what is available to the user:

# To install previous version of wsrf without C++11R> install.packages("wsrf",type="source",configure.args="--enable-c11=no")# To install previous version of wsrf with Boost for multithreadingR> install.packages("wsrf",+type="source",+configure.args="--with-boost-include=<Boost include path>                                      --with-boost-lib=<Boost lib path>")

After installation, one can use the built-in functionwsrfParallelInfo to query whether the version installed is what theyreally want (distributed or multi-threaded).

R> wsrfParallelInfo()

GPL (>= 2)