skglm.datafits.QuadraticHessian

class skglm.datafits.QuadraticHessian

Quadratic datafit where we pass the Hessian A directly.

The datafit reads:

`1 / 2 x^(\top) A x + \langle b, x \rangle`

For a symmetric A. Up to a constant, it is the same as a Quadratic, with `A = 1 / (n_"samples") X^(\top)X` and `b = - 1 / n_"samples" X^(\top)y`. When the Hessian is available, this datafit is more efficient than using Quadratic.

__init__()

Methods

__init__()
get_lipschitz(A, b)
get_spec()	Specify the numba types of the class attributes.
gradient_scalar(A, b, w, Ax, j)
initialize(X, y)	Pre-computations before fitting on X and y.
initialize_sparse(X_data, X_indptr, X_indices, y)	Pre-computations before fitting on X and y when X is a sparse matrix.
params_to_dict()	Get the parameters to initialize an instance of the class.
value(b, x, Ax)	Value of datafit at vector w.