Transformation to Other Sparse Formats

Transform between the spam sparse format to the matrix.csr format of SparseM and dgRMatrix format of Matrix

as.spam.matrix.csr(x)
as.dgRMatrix.spam(x)
as.dgCMatrix.spam(x)
as.spam.dgRMatrix(x)
as.spam.dgCMatrix(x)

Arguments

x

sparse matrix of class spam, matrix.csr, dgRMatrix or dgCMatrix.

Value

According to the call, a sparse matrix of class spam,

matrix.csr, dgRMatrix or dgCMatrix.

Details

We do not provide any S4 methods and because of the existing mechanism a standard S3 does not work.

The functions are based on require.

Notice that as.matrix.csr.spam should read as as."matrix.csr".spam.

See also

triplet, Matrix or matrix.csr from package SparseM.

Author

Reinhard Furrer

Examples

if (FALSE) {
S <- diag.spam(4)
R <- as.dgRMatrix.spam( S)
C <- as.dgCMatrix.spam( S)
as.spam.dgCMatrix(C)
slotNames(C)
slotNames(R)
# For column oriented sparse formats a transpose does not the job,
# as the slot names change.


# as.spam(R) does not work.
}

if (FALSE) {
# for transformations between SparseM and spam:

as.matrix.csr.spam <- function(x,...) {
  if (new("matrix.csr")) {
    newx <- new("matrix.csr")
    slot(newx,"ra",check=FALSE) <- x@entries
    slot(newx,"ja",check=FALSE) <- x@colindices
    slot(newx,"ia",check=FALSE) <- x@rowpointers
    slot(newx,"dimension",check=FALSE) <- x@dimension
    return(newx)
    }
  stop("function requires 'SparseM' package")
}
# then with `SparseM`:  as.matrix.csr.spam( spamobject )
}

if (FALSE) {
# a dataset contained in Matrix
data(KNex, package='Matrix')
summary( KN <- as.spam.dgCMatrix(KNex$mm) )

}