# pmclust v0.2-0

0

0th

Percentile

## Parallel Model-Based Clustering using Expectation-Gathering-Maximization Algorithm for Finite Mixture Gaussian Model

Aims to utilize model-based clustering (unsupervised) for high dimensional and ultra large data, especially in a distributed manner. The code employs 'pbdMPI' to perform a expectation-gathering-maximization 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 programming model. The code can be executed through 'pbdMPI' and MPI' implementations such as 'OpenMPI' and 'MPICH'. See the High Performance Statistical Computing website <https://snoweye.github.io/hpsc/> for more information, documents and examples.

## Functions in pmclust

 Name Description One E-Step Compute One E-step and Log Likelihood Based on Current Parameters generate.basic Generate Examples for Testing Set of PARAM A Set of Parameters in Model-Based Clustering. Internal Functions All Internal Functions One Step of EM algorithm One EM Step for GBD One M-Step Compute One M-Step Based on Current Posterior Probabilities assign.N.sample Obtain a Set of Random Samples for X.spmd Independent logL Independent Function for Log Likelihood pmclust-package Parallel Model-Based Clustering Update Class of EM or Kmenas Results Update CLASS.spmd Based on the Final Iteration Set of CONTROL A Set of Controls in Model-Based Clustering. mb.print Print Results of Model-Based Clustering pmclust and pkmeans Parallel Model-Based Clustering and Parallel K-means Algorithm print.object Functions for Printing or Summarizing Objects According to Classes as functions Convert between X.gbd (X.spmd) and X.dmat get.N.CLASS Obtain Total Elements for Every Clusters Initialization Initialization for EM-like Algorithms EM-like algorithms EM-like Steps for GBD Read Me First Read Me First Function Set Global Variables Set Global Variables According to the global matrix X.gbd (X.spmd) or X.dmat generate.MixSim Generate MixSim Examples for Testing No Results!

## Vignettes of pmclust

 Name pmclust-include/00-preamble.tex pmclust-include/01-acknowledgement.tex pmclust-include/01-introduction.tex pmclust-include/02-example.tex pmclust-include/03-algorithm.tex pmclust-include/04-discussion.tex pmclust-include/my_jss.cls pmclust-include/pmclust.bib build_pdf.sh pmclust-guide.Rnw No Results!