| ddmatrix-apply {pbdDMAT} | R Documentation |
Apply a function to the margins of a distributed matrix.
## S4 method for signature 'ddmatrix' apply(X, MARGIN, FUN, ..., reduce = FALSE, proc.dest = "all")
X |
distributed matrix |
MARGIN |
subscript over which the function will be applied |
FUN |
the function to be applied |
... |
additional arguments to FUN |
reduce |
logical or string. See details |
proc.dest |
Destination process (or 'all') if a reduction occurs |
If reduce==TRUE then a global matrix or vector (whichever is more
appropriate) will be returned. The argument proc.dest= behaves
exactly as in the as.vector() and as.matrix() functions of
pbdDMAT. If reduce=FALSE then a distributed matrix is returned.
Other acceptable arguments are reduce="matrix" and
reduce="vector" which demand global matrix or vector return,
respectively. This should generally be slightly more efficient than running
apply and then calling as.vector() or as.matrix().
Returns a distributed matrix unless a reduction is requested, then a global matrix/vector is returned.
## Not run: # Save code in a file "demo.r" and run with 2 processors by # > mpiexec -np 2 Rscript demo.r library(pbdDMAT, quiet = TRUE) init.grid() x <- ddmatrix(1:9, 3, bldim=2) y <- as.vector(apply(x, 1, mean)) comm.print(y) finalize() ## End(Not run)