pyriemann.utils.mean.mean_logdet

pyriemann.utils.mean.mean_logdet(covmats, tol=0.0001, maxiter=50, init=None, sample_weight=None)

Mean of SPD matrices according to the log-det metric.

Log-det mean is obtained by an iterative procedure where the update is:

\[\mathbf{C} = \left(\sum_i \left( 0.5 \mathbf{C} + 0.5 \mathbf{C}_i \right)^{-1} \right)^{-1}\]
Parameters
covmatsndarray, shape (n_matrices, n_channels, n_channels)

Set of SPD matrices.

tolfloat, default=10e-5

The tolerance to stop the gradient descent.

maxiterint, default=50

The maximum number of iterations.

initNone | ndarray, shape (n_channels, n_channels), default=None

A SPD matrix used to initialize the gradient descent. If None, the weighted Euclidean mean is used.

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

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

Returns
Cndarray, shape (n_channels, n_channels)

Log-det mean.