colAvgsPerRowSet,xgCMatrix-method {sparseMatrixStats} | R Documentation |
Calculates for each row (column) a summary statistic for equally sized subsets of columns (rows)
## S4 method for signature 'xgCMatrix' colAvgsPerRowSet( X, W = NULL, cols = NULL, S, FUN = colMeans2, ..., na.rm = NA, tFUN = FALSE ) ## S4 method for signature 'xgCMatrix' rowAvgsPerColSet( X, W = NULL, rows = NULL, S, FUN = rowMeans2, ..., na.rm = NA, tFUN = FALSE )
X |
An |
W |
An optional numeric |
cols |
A |
S |
An integer |
FUN |
A row-by-row (column-by-column) summary statistic function. It is
applied to to each column (row) subset of |
... |
Additional arguments passed to |
na.rm |
(logical) Argument passed to |
tFUN |
If |
rows |
A |
**Note**: the handling of missing parameters differs from [matrixStats::colAvgsPerRowSet()]. The 'matrixStats' version always removes ‘NA'’s if there are any in the data. This method however does whatever is passed in the '...' parameter.
Returns a numeric JxN
(MxJ
) matrix.
matrixStats::rowAvgsPerColSet()
and matrixStats::colAvgsPerRowSet()
which are used when the input is a matrix
or numeric
vector.
mat <- matrix(rnorm(20), nrow = 5, ncol = 4) mat[2, 1] <- NA mat[3, 3] <- Inf mat[4, 1] <- 0 print(mat) S <- matrix(1:ncol(mat), ncol = 2) print(S) rowAvgsPerColSet(mat, S = S, FUN = rowMeans) rowAvgsPerColSet(mat, S = S, FUN = rowVars)