covmats.LinearDriftMatrix#

class covmats.LinearDriftMatrix(pts: ndarray[tuple[Any, ...], dtype[floating]])[source]#

Represent a linear (order 1) drift matrix (trend).

The drift basis is made of a constant column followed by the raw point coordinates, \(\mathbf{X} = [\mathbf{1}, \mathbf{pts}]\), with shape \((N_s, 1 + N_{dim})\), so that \(\mathbf{X}\boldsymbol{\beta}\) represents a mean varying linearly with the spatial coordinates (i.e., universal kriging with a linear trend).

__init__(pts: ndarray[tuple[Any, ...], dtype[floating]]) → None[source]#

Initialize the instance.

Parameters:: pts (NDArrayFloat) – Coordinates of the points, with shape (Ns, Ndim), Ndim being the number of spatial dimensions.

Properties

`mat`
`beta_dim`
`s_dim`

Methods

`dot`	Return the dot product.
`get_gradient_dot_product`	Return the dot product of the gradient of the prior and the given input vector.
`get_values`	Return the values of the prior term.