Learn R Programming
⚠️There's a newer version (0.1.8) of this package.Take me there.

bigGP (version 0.1-5)

Distributed Gaussian Process Calculations

Description

Distributes Gaussian process calculations across nodes in a distributed memory setting, using Rmpi. The bigGP class provides high-level methods for maximum likelihood with normal data, prediction, calculation of uncertainty (i.e., posterior covariance calculations), and simulation of realizations. In addition, bigGP provides an API for basic matrix calculations with distributed covariance matrices, including Cholesky decomposition, back/forwardsolve, crossproduct, and matrix multiplication.

Copy Link

Version

Install

install.packages('bigGP')

Monthly Downloads

102

Version

0.1-5

License

GPL (>= 2)

Maintainer

Christopher Paciorek

Last Published

January 21st, 2015

Functions in bigGP (0.1-5)

remoteLs

Remote List Objects
distributeVector

Distribute a Vector to the Slave Processes
remoteCrossProdMatVec

Distributed Crossproduct of a Rectangular Matrix and a Vector
remoteCalc

Return a Distributed Vector to the Master Process
localKrigeProblemConstructMean

Calculate Mean Vector or Covariance Matrix on Slave Process
distributedKrigeProblem-class

ReferenceClass for Distributed Components of the krigeProblem ReferenceClass
remoteRm

Remote Remove Objects
bigGP.init

Initialize bigGP package
alloc

Create Object with its Own Memory
remoteCrossProdMatSelf

Distributed Crossproduct of a Rectangular Matrix with Itself
localCalc

Local Calculation Functions
localCollectVector

Local Distribution and Collection Functions
getDistributedVectorLength

Find Length of Subset of Vector or Matrix Stored on Slave Process
SN2011fe

SN2011fe Supernova Dataset
remoteGetIndices

Determine Indices of Vector or Matrix Elements Stored on all Processes
collectDiagonal

Return the Diagonal of a Distributed Square Matrix to the Master Process
push

Copy Object from Master to Slave Processes
pull

Copy Object from Slave Processes to Master
localAssign

Assign a New Name to an Object on Slave Process
collectRectangularMatrix

Return a Distributed Rectangular Matrix to the Master Process
remoteMultChol

Distributed Multiplication of Lower Triangular Matrix and a Vector or Matrix
bigGP

Package for Calculations with Big Gaussian Processes
krigeProblem-class

Class "krigeProblem"
collectVector

Return a Distributed Vector to the Master Process
calcD

Calculate Partition Factor
remoteForwardsolve

Solve a Distributed Triangular System
collectTriangularMatrix

Return a Distributed Symmetric or Triangular Matrix to the Master Process
calcIJ

Calculate Slave Process Identifiers
localRm

Remove Objects on Slave Process
bigGP.exit

Exit bigGP Environment
bigGP-meta

Information about the number and identities of the processes
localGetVectorIndices

Get Indices of Vector or Matrix Elements Stored on Slave Process
remoteConstructRnormVector

Create Distributed Vector or Matrix of Random Normals
remoteCalcChol

Calculate Distributed Cholesky Decomposition