bigGP (version 0.1-5)

collectRectangularMatrix: Return a Distributed Rectangular Matrix to the Master Process

Description

collectRectangularMatrix retrieves a distributed rectangular matrix from the slave processes, reconstructing the blocks correctly on the master process. Objects can be copied from environments, lists, and ReferenceClass objects as well as the global environment on the slave processes. WARNING: do not use with a distributed symmetric square matrix; instead use collectTriangularMatrix.

Usage

collectRectangularMatrix(objName, objPos = '.GlobalEnv', n1, n2, h1 = 1, h2 = 1)

Value

collectRectangularMatrix returns a matrix of dimension, $n1 \times n2$.

See Also

pull collectVector collectTriangularMatrix collectDiagonal distributeVector

Examples

if(require(fields)) {
nProc <- 3
n <- nrow(SN2011fe_subset)
m <- nrow(SN2011fe_newdata_subset)
inputs <- c(as.list(SN2011fe_subset), as.list(SN2011fe_newdata_subset),
  nu =2)
# initialize the problem
prob <- krigeProblem$new("prob", h_n = 1, h_m = 1, numProcesses =
  nProc, n = n, m = m,
  meanFunction = SN2011fe_meanfunc, predMeanFunction = SN2011fe_predmeanfunc, 
  covFunction = SN2011fe_covfunc, crossCovFunction = SN2011fe_crosscovfunc,
predCovFunction = SN2011fe_predcovfunc, params = SN2011fe_mle$par,
  inputs = inputs, data = SN2011fe_subset$flux, packages = c("fields"))
# do predictions, primarily so cross-covariance gets calculated
pred <- prob$predict(ret = TRUE, verbose = TRUE)

crossC <- collectRectangularMatrix('crossC', "prob", n1 = n, n2 = m,
h1 = 1, h2 = 1)
crossC[1:5, 1:5]
}