Recommend Practical Defaults for Bilinear Mixed

Provides data-adaptive defaults for [bilinear_mixed()] so users can avoid manual specification of many tuning parameters.

Usage

bilinear_mixed_recommend(
  data,
  subject,
  z = NULL,
  y = NULL,
  row_design = NULL,
  mode = c("auto", "seed_axis", "seed_repeat", "both"),
  connectivity_type = c("auto", "cross", "symmetric"),
  profile = c("fast", "balanced", "adaptive"),
  max_r_seed = 6,
  max_r_roi = 20,
  max_k_subject = 5,
  sym_tol = 1e-08
)

Arguments

data

A list of numeric connectivity matrices.

subject

Subject identifier with length length(data).

z

Optional repeat-level design.

y

Optional subject-level traits.

row_design

Optional row-level covariates (same number of rows as each connectivity matrix).

mode

One of "auto", "seed_axis", "seed_repeat", or "both".

connectivity_type

One of "auto", "cross", or "symmetric".

profile

One of "fast", "balanced", or "adaptive".

max_r_seed

Maximum suggested row rank.

max_r_roi

Maximum suggested column rank.

max_k_subject

Maximum suggested subject latent dimension.

sym_tol

Numeric tolerance for symmetry detection when connectivity_type = "auto".

Value

Named list of recommended arguments for [bilinear_mixed()].