runamica

runamica(data, num_models=1, max_iter=2000, num_mix_comps=3, pcakeep=None, outdir=None, amica_binary=None, max_threads=4, cleanup=True, **kwargs)

Run AMICA binary on a data matrix.

Parameters

datandarray, shape (chans, frames)

Input data matrix (float64).

num_modelsint

Number of ICA models to fit.

max_iterint

Maximum number of training iterations.

num_mix_compsint

Number of mixture components per source.

pcakeepint or None

Number of principal components to retain. Default: chans.

outdirstr or None

Output directory. If None, a temporary directory is created.

amica_binarystr or None

Path to AMICA binary. If None, auto-detected.

max_threadsint

Maximum number of threads for the binary.

cleanupbool

If True, remove temporary files after loading results.

**kwargs

Additional AMICA parameters written to the param file as-is.

Returns

weightsndarray, shape (num_pcs, num_pcs)

Unmixing weights for model 0 (post-sphering).

spherendarray, shape (num_pcs, chans)

Sphering matrix.

modsdict

Full model output with keys: W, S, A, mean, gm, c, alpha, mu, sbeta, rho, LL, LLt, nd, comp_list, svar, origord, mod_prob, num_pcs, etc.