tedana.decomposition.ica.r_ica

r_ica(data, n_components, fixed_seed, n_robust_runs, max_it, n_threads=1)[source]

Perform robustica on data and returns mixing matrix.

Parameters:

  • data ((Mc x T) numpy.ndarray) – Dimensionally reduced optimally combined functional data, where S is samples in classification mask, T is time

  • n_components (int) – Number of components retained from PCA decomposition.

  • fixed_seed (int) – Seed for ensuring reproducibility of ICA results.

  • n_robust_runs (int) – selected number of robust runs when robustica is used. Default is 30.

  • maxit (int, optional) – Maximum number of iterations for ICA. Default is 500.

  • n_threads (int, optional) – Number of threads to use for parallel computation. Default is 1.

Returns:

  • mixing ((T x C) numpy.ndarray) – Z-scored mixing matrix for converting input data to component space, where C is components and T is the same as in data

  • fixed_seed (int) – Random seed from final decomposition.

  • c_labels ((n_pca_components x n_robust_runs,) numpy.ndarray) – A one dimensional array that has the cluster label of each run.

  • similarity_t_sne ((n_pca_components x n_robust_runs,2) numpy.ndarray) – An array containing the 2D coordinates of projected data.

  • fastica_convergence_warning_count (int) – The number of iterations of fastICA that failed to converge.

  • index_quality (float) – The mean cluster index quality for robustICA. robustICA cites https://doi.org/10.1109/NNSP.2003.1318025 for the measure