pymc.gp.cov.Matern52

class pymc.gp.cov.Matern52(input_dim, ls=None, ls_inv=None, active_dims=None)

The Matérn kernel with \(\nu = \frac{5}{2}\).

\[k(x, x') = \left(1 + \frac{\sqrt{5(x - x')^2}}{\ell} + \frac{5(x-x')^2}{3\ell^2}\right) \mathrm{exp}\left[ - \frac{\sqrt{5(x - x')^2}}{\ell} \right]\]

Read more here.

Parameters:

input_dimint

The number of input dimensions

lsscalar or array, optional

Lengthscale parameter \(\ell\); if input_dim > 1, a list or array of scalars. If input_dim == 1, a scalar.

ls_invscalar or array, optional

Inverse lengthscale \(1 / \ell\). One of ls or ls_inv must be provided.

active_dimslist of int, optional

The dimension(s) the covariance function operates on.

Methods

Matern52.__init__(input_dim[, ls, ls_inv, ...])
Matern52.diag(X)
Matern52.euclidean_dist(X, Xs)
Matern52.full(X[, Xs])
Matern52.full_from_distance(dist[, squared])
Matern52.power_spectral_density(omega)	Power spectral density for the Matern52 kernel.
Matern52.square_dist(X, Xs)

Attributes

n_dims

The dimensionality of the input, as taken from the active_dims.