eztaox.kernels.direct#

Kernels evaluated using a direct apporach.

The likelihood computation follows O(N^3) scaling.

Classes#

`MultibandLowRank`	A multiband kernel implementating a low-rank Kronecker covariance structure.
`MultibandFullRank`	A multiband kernel implementating the full-rank Kronecker covariance structure.

Module Contents#

class MultibandLowRank[source]#

Bases: tinygp.kernels.Kernel

A multiband kernel implementating a low-rank Kronecker covariance structure.

The specific form of the cross-band Kronecker covariance matrix is given by Equation 13 of Gordon et al. (2020). The implementation is inspired by this tinygp tutorial.

coord_to_sortable(X) → tinygp.helpers.JAXArray[source]#: Extract the time-sortable component of the coordinates.

evaluate(X1, X2) → tinygp.helpers.JAXArray[source]#: Evaluate the kernel at a pair of input coordinates.

class MultibandFullRank(kernel, diagonal, off_diagonal)[source]#

Bases: tinygp.kernels.Kernel

A multiband kernel implementating the full-rank Kronecker covariance structure.

The specific form of the cross-band Kronecker covariance matrix is given by Equation 18-20 of Gordon et al. (2020). The implementation is inspired by this tinygp tutorial.

Note

This kernel is still in development, please use with caution.

coord_to_sortable(X) → tinygp.helpers.JAXArray[source]#: Extract the time-sortable component of the coordinates.

evaluate(X1, X2) → tinygp.helpers.JAXArray[source]#: Evaluate the kernel at a pair of input coordinates.