pyeeg.solvers.ScipyRobustSolver.solve

ScipyRobustSolver.solve(X, y, alpha=0.0, M=None)

Fit the robust Cauchy-loss regression and return coefficients.

Fits each output channel with scipy.optimize.least_squares using the 'cauchy' loss. Only unregularized regression is supported: alpha and M are accepted for API compatibility but ignored.

Parameters:

  • X (ndarray or list of ndarray) – Design matrix (n_samples, n_features) or list of segments.

  • y (ndarray or list of ndarray) – Target (n_samples, n_channels) or list of arrays (one per segment).

  • alpha (float or array-like, optional) – Accepted for API compatibility; not used by this solver. Default is 0.0.

  • M (ndarray, optional) – Accepted for API compatibility; not used by this solver. Default is None.

Returns:

result – betas has shape (n_features, n_channels). info is a dict with keys n_iter (max nfev across channels), converged (True if all channels succeeded), scale (array of per-channel scales), and results (list of scipy.optimize.OptimizeResult objects).

Return type:

SolverResult