create.numeric.grad: Create a numerical gradient function for a kernel using the...

In mattdneal/gaussianProcess: Gaussian Process Fitting and Prediction

Description

Create a numerical gradient function for a kernel using the grad function.

Usage

create.numeric.grad(k, method = "Richardson", method.args = NULL,
  additional.params = list())

Arguments

k	kernel function
method	method for creating numeric grad, see grad
method.args	method arguments for creating numeric grad, see grad
additional.params	additional parameters for the kernel function k

Value

A function which gives the numeric grad of k w.r.t. the kernel hyperparameters.

mattdneal/gaussianProcess documentation built on May 21, 2019, 12:58 p.m.