pyriemann.utils.mean.mean_covariance(X=None, metric='riemann', sample_weight=None, covmats=None, **kwargs)

Mean of matrices according to a metric.

Compute the mean of a set of matrices according to a metric [1].

Xndarray, shape (n_matrices, n, n)

Set of matrices.

metricstring | callable, default=”riemann”

Metric for mean estimation, can be: “ale”, “alm”, “euclid”, “harmonic”, “identity”, “kullback_sym”, “logdet”, “logeuclid”, “riemann”, “wasserstein”, or a callable function.

sample_weightNone | ndarray, shape (n_matrices,), default=None

Weights for each matrix. If None, it uses equal weights.


The keyword arguments passed to the sub function.

Mndarray, shape (n, n)

Mean of matrices.



Review of Riemannian distances and divergences, applied to SSVEP-based BCI S. Chevallier, E. K. Kalunga, Q. Barthélemy, E. Monacelli. Neuroinformatics, Springer, 2021, 19 (1), pp.93-106