Apply a function to bounded feature blocks

Usage

block_apply(x, FUN, block_size = 4096L, assay = active_assay(x), ...)

Arguments

x: An fmri_frame or view.
FUN: Function receiving an observation-by-feature matrix and feature IDs.
block_size: Number of features per block.
assay: Assay name.
...: Additional arguments passed to FUN.

Value

A list of block results.

Examples

sp <- volume_space(dim = c(2, 2, 2), affine = diag(4))
frame <- fmri_frame(
  assays = list(bold = matrix(rnorm(4 * n_features(sp)), nrow = 4)),
  observations = data.frame(.obs_id = sprintf("vol-%d", 1:4)),
  space = sp
)
block_apply(frame, function(mat, ids) colMeans(mat), block_size = 4L)
#> [[1]]
#> [1]  0.4030433 -0.7562311 -0.8855047 -0.1006778
#> 
#> [[2]]
#> [1] 0.62804704 0.01570424 1.07680091 0.11770979
#>