dagma.linear.DagmaLinear._func¶

dagma.linear.DagmaLinear._func(W: numpy.ndarray, mu: float, s: float = 1.0) → tuple[float, numpy.ndarray]¶

Evaluate value of the penalized objective function.

Parameters:¶

W : np.ndarray¶

\((d,d)\) adjacency matrix

mu : float¶

Weight of the score function.

s : float, optional¶

Controls the domain of M-matrices. Defaults to 1.0.

Returns:¶

Objective value, and gradient of the objective

Return type:¶

Tuple[float, np.ndarray]