Skip to contents

Fast ICA via Rust

fast_ica_rust.Rd

This functions is a wrapper over the Rust implementation of fastICA and the generation of the pre-processed data and pre-whitening matrix. It has the same two options c("logcosh", "exp") to run ICA in parallel modus. You can control the parameters of ICA via ica_params.

Usage

fast_ica_rust(
  X,
  n_icas,
  ica_fun = c("logcosh", "exp"),
  ica_params = params_ica_general(),
  fast_svd = TRUE,
  seed = NULL
)

Arguments

X

Numeric matrix. The data on which you want to run fastICA.

n_icas

Integer. Number of independent components to recover.

ica_fun

String, element of c("logcosh", "exp").

ica_params

List. The ICA parameters, see params_ica_general() wrapper function. This function generates a list containing:

  • maxit - Integer. Maximum number of iterations for ICA.

  • alpha - Float. The alpha parameter for the logcosh version of ICA. Should be between 1 to 2.

  • max_tol - Maximum tolerance of the algorithm.

  • verbose - Controls verbosity of the function.

fast_svd

Boolean. Shall the randomised SVD be used. This is faster, but less precise.

seed

Integer. Seed to ensure reproducible results.

Value

A list containing:

  • w The mixing matrix w.

  • A ICA results matrix A.

  • S ICA results matrix S.

  • converged Boolean indicating if algorithm converged.

Examples

# recover two mixed sources
sources <- cbind(sin((1:1000) / 20), rep(((1:200) - 100) / 100, 5))
mixed <- sources %*% matrix(c(0.291, 0.6557, -0.5439, 0.5572), 2, 2)
ica_res <- fast_ica_rust(mixed, n_icas = 2L, ica_fun = "logcosh", seed = 42L)
max(abs(cor(sources[, 1], t(ica_res$S))))
#> [1] 0.9999303