Consider this simple sparse matrix
> (X <- sparseMatrix(c(1, 2, 1), c(1, 1, 2), x = 0:2))
2 x 2 sparse Matrix of class "dgCMatrix"
[1,] 0 2
[2,] 1 .
How can I convert it into a matrix indicating if the corresponding element is non-empty? Here is what I'm doing now, but being 0 does not equal to being "empty", and this approach doesn't differentiate between them.
> (Y <- X != 0)
2 x 2 sparse Matrix of class "lgCMatrix"
[1,] : |
[2,] | .
To clarify, the desired output may only contain TRUE or FALSE but not NA. It can be either a matrix or a sparseMatrix. Even more preferably, it may be a list, in which case each slot corresponds to a column of X. For example, the answer for X should be either
[,1] [,2]
[1,] TRUE TRUE
[2,] TRUE FALSE
or
$`1`
[1] TRUE TRUE
$`2`
[1] TRUE FALSE
pex <- function(p) {
dp <- diff(p)
rep(seq_along(dp), dp)
}
m = matrix(FALSE, nrow = nrow(X), ncol = ncol(X))
m[cbind(X@i + 1, pex(X@p))] = TRUE
m
# [,1] [,2]
#[1,] TRUE TRUE
#[2,] TRUE FALSE
pex is from here.