gp.log.marg.like: The log marginal likelihood. See "2.3 Varying the Hyperparameters"
on page 19 of Rasmumssen and Williams' book.
Description
The log marginal likelihood. See "2.3 Varying the Hyperparameters"
on page 19 of Rasmumssen and Williams' book.
Usage
gp.log.marg.like(y, K = NULL, U = chol(K))
Arguments
K
The covariance matrix (kernel), not needed if U is provided.
U
Cholesky decomposition of K (chol(K)).