| CsparseMatrix-class {Matrix} | R Documentation |
The "CsparseMatrix" class is the virtual class of
all sparse matrices coded in sorted compressed column-oriented form.
Since it is a virtual class, no objects may be created from it. See
showClass("CsparseMatrix") for its subclasses.
i:"integer" of length nnzero
(number of non-zero elements). These are the 0-based row numbers for
each non-zero element in the matrix, i.e., i must be in
0:(nrow(.)-1)p:"integer" for providing pointers, one
for each column, to the initial (zero-based) index of elements in
the column. .@p is of length ncol(.) + 1, with
p[1] == 0 and p[length(p)] == nnzero, such that in
fact, diff(.@p) are the number of non-zero elements for
each column.Dim, Dimnames:sparseMatrix class.
Class "sparseMatrix", directly.
Class "Matrix", by class "sparseMatrix".
signature(e1 = "CsparseMatrix", e2 = "numeric"): ... signature(e1 = "numeric", e2 = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(e1 = "CsparseMatrix", e2 = "numeric"): ... signature(e1 = "numeric", e2 = "CsparseMatrix"): ... signature(x = "CsparseMatrix", i = "index", j = "missing", value = "replValue"): ... signature(x = "CsparseMatrix", i = "index", j = "index", value = "replValue"): ... signature(x = "CsparseMatrix", i = "missing", j = "index", value = "replValue"): ... signature(x = "CsparseMatrix", y = "CsparseMatrix"): ... signature(x = "CsparseMatrix", y = "denseMatrix"): ... signature(x = "CsparseMatrix", y = "matrix"): ... signature(e1 = "CsparseMatrix", e2 = "numeric"): ... signature(e1 = "numeric", e2 = "CsparseMatrix"): ... signature(from = "CsparseMatrix", to = "TsparseMatrix"): ... signature(from = "CsparseMatrix", to = "denseMatrix"): ... signature(from = "CsparseMatrix", to = "matrix"): ... signature(from = "CsparseMatrix", to = "lsparseMatrix"): ... signature(from = "CsparseMatrix", to = "nsparseMatrix"): ... signature(from = "TsparseMatrix", to = "CsparseMatrix"): ... signature(from = "denseMatrix", to = "CsparseMatrix"): ... signature(x = "CsparseMatrix", y = "CsparseMatrix"): ... signature(x = "CsparseMatrix", y = "missing"): ... signature(x = "CsparseMatrix", y = "dgeMatrix"): ... signature(x = "CsparseMatrix", y = "matrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix", y = "CsparseMatrix"): ... signature(x = "CsparseMatrix", y = "missing"): ... signature(x = "CsparseMatrix"): ... signature(x = "CsparseMatrix"): ...
All classes extending CsparseMatrix have a common validity
(see validObject) check function. That function
additionally checks the i slot for each column to contain
increasing row numbers.
In earlier versions of Matrix (<= 0.999375-16),
validObject automatically re-sorted the entries when
necessary, and hence new() calls with somewhat permuted
i and x slots worked, as new(...)
(with slot arguments) automatically checks the validity.
Now, you have to use sparseMatrix to achieve the same
functionality or know how to use .validateCsparse() to do so.
colSums, kronecker, and other such methods
with own help pages.
Further, the super class of CsparseMatrix,
sparseMatrix, and, e.g.,
class dgCMatrix for the links to other classes.
getClass("CsparseMatrix")
## The common validity check function (based on C code):
getValidity(getClass("CsparseMatrix"))
## This behavior has been changed
# un-sorted i-slot on input is automatically "sorted" :
#M <- new("dgCMatrix", Dim = 2:3, p = as.integer(c(0,1,1,3)),
# i = c(1L, 1:0), x= c(2,2,1))
#stopifnot(M@i == c(1L, 0:1), M@x == c(2,1,2))