mne_icalabel.iclabel.get_iclabel_features

mne_icalabel.iclabel.get_iclabel_features(inst, ica)

Generate the features for ICLabel neural network.

Parameters:

instRaw | Epochs

MNE Raw/Epoch instance with data array in Volts.

icaICA

MNE ICA decomposition.

Returns:

topoarray of shape (32, 32, 1, n_components)

The topoplot feature.

psdarray of shape (1, 100, 1, n_components)

The psd feature.

autocorrarray of shape (1, 100, 1, n_components)

The autocorrelations feature. Depending on the length of the raw data passed in, different methods of computing autocorrelation will be used. See Pion-Tonachini et al.[1] for details.

References

[1]

Luca Pion-Tonachini, Ken Kreutz-Delgado, and Scott Makeig. Iclabel: an automated electroencephalographic independent component classifier, dataset, and website. NeuroImage, 198:181–197, September 2019. doi:10.1016/j.neuroimage.2019.05.026.