pyeeg.models.TRFEstimator.multialpha_score

TRFEstimator.multialpha_score(X, y)

Score the fitted multi-alpha models against held-out data.

Computes the per-channel correlation between the data predicted by the coefficients stored in all_betas (one solution per alpha) and the true target y. The estimator must have been fitted with an array of several alpha values at once.

Parameters:

• X (ndarray or list of ndarray) – Feature matrix (already lagged or not). A list of segments is accepted and must match y in length; scores are averaged across segments.

• y (ndarray or list of ndarray) – EEG data, shape (nsamples, nchans) for a single subject or (nsubjects, nsamples, nchans) for several subjects.

Returns:

scores – Array of shape (nsegments_or_subjects, n_alphas, n_chans) containing the per-channel correlation score for each alpha.

Return type:

ndarray