EEG processing pipeline with autoreject

[richard (Richard Höchenberger)]

@eduardo

epochs_clean = epochs.copy()
ica.apply(epochs_clean)

Since apply() returns the modified instance, you could simplify this to:

epochs_clean = ica.apply(epochs.copy())

@eduardo

My question is, do I also need to standarize the EEG epochs before extracting the features?

The way you’re phrasing it now, I would say: no. Just be sure to standardize whatever you feed into the ML algorithm.