spatial_constraints: spatial constraints

In bbuchsbaum/neuropls: neuroca: multiblock component analysis for neuroimaging data

spatial constraints

Description

construct a sparse matrix of spatial constraints for a set of blocks

Usage

spatial_constraints(
  coords,
  nblocks = 1,
  sigma_within = 5,
  sigma_between = 1,
  shrinkage_factor = 0.1,
  nnk_within = 27,
  nnk_between = 1,
  weight_mode_within = "heat",
  weight_mode_between = "binary",
  variable_weights = rep(1, ncol(coords) * nblocks),
  verbose = FALSE
)

Arguments

coords	the spatial coordinates as a matrix with rows as objects and columns as dimensions
nblocks	the number of repeated coordinate blocks
sigma_within	the bandwidth of the within-block smoother
sigma_between	the bandwidth of the between-block smoother
shrinkage_factor	the amount of shrinkage towards the spatially localized average
nnk_within	the maximum number of nearest neighbors for within block smoother
nnk_between	the maximum number of nearest neighbors for between block smoother

bbuchsbaum/neuropls documentation built on April 17, 2022, 8:46 a.m.