pmclust: Parallel Model-Based Clustering

The pmclust aims to utilize model-based clustering (unsupervised) for high dimensional and ultra large data, especially in a distributed manner. The package employs Rmpi to perform a expectation-gathering-maximization (EGM) algorithm for finite mixture Gaussian models. The unstructured dispersion matrices are assumed in the Gaussian models. The implementation is default in the single program multiple data (SPMD) programming model. The code can be executed through Rmpi and independent to most MPI applications. See the High Performance Statistical Computing (HPSC) website for more information, documents and examples.

Version: 0.1-6
Depends: R (≥ 3.0.0), methods, rlecuyer, pbdMPI (≥ 0.2-2), MASS
Enhances: MixSim, pbdSLAP (≥ 0.1-7), pbdBASE (≥ 0.3-0), pbdDMAT (≥ 0.2-4)
Published: 2014-02-03
Author: Wei-Chen Chen [aut, cre], George Ostrouchov [aut]
Maintainer: Wei-Chen Chen <wccsnow at>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
Citation: pmclust citation info
Materials: README ChangeLog
In views: Cluster, HighPerformanceComputing
CRAN checks: pmclust results


Reference manual: pmclust.pdf
Vignettes: pmclust-guide
Package source: pmclust_0.1-6.tar.gz
Windows binaries: r-devel: not available, r-release: not available, r-oldrel: not available
OS X Snow Leopard binaries: r-release: pmclust_0.1-6.tgz, r-oldrel: pmclust_0.1-6.tgz
OS X Mavericks binaries: r-release: not available
Old sources: pmclust archive

Reverse dependencies:

Reverse enhances: pbdDEMO