Scale a delayed matrix along rows or columns

Usage

d_scale(x, dim = c("rows", "cols"), center = TRUE, scale = TRUE, na.rm = FALSE)

Arguments

x: A delarr.
dim: Dimension to scale.
center: Logical; subtract the mean before scaling.
scale: Logical; divide by the standard deviation.
na.rm: Logical; remove missing values when computing statistics.

Value

A delarr with a deferred scaling operation.

Examples

mat <- matrix(c(1, 2, 3, 10, 20, 30), nrow = 2, ncol = 3)
darr <- delarr(mat)

# Scale rows (center and divide by SD)
scaled <- darr |> d_scale(dim = "rows") |> collect()
scaled
#>            [,1]       [,2]     [,3]
#> [1,] -0.6704784 -0.4789131 1.149392
#> [2,] -0.8320503 -0.2773501 1.109400

# Scale without centering
scaled_only <- darr |> d_scale(dim = "rows", center = FALSE) |> collect()
scaled_only
#>            [,1]      [,2]     [,3]
#> [1,] 0.09578263 0.2873479 1.915653
#> [2,] 0.13867505 0.6933752 2.080126