Kernel: The 'Kernel' class object

In KGode: Kernel Based Gradient Matching for Parameter Inference in Ordinary Differential Equations

The 'Kernel' class object

Description

This a abstract class provide the kernel function and the 1st order derivative of rbf kernel function.

Format

R6Class object.

Value

an R6Class object which can be used for the rkhs interpolation.

Methods

kern(t1,t2)

This method is used to calculate the kernel function given two one dimensional real inputs.

dkd_kpar(t1,t2)

This method is used to calculate the gradient of kernel function against the kernel hyper parameters given two one dimensional real inputs.

dkdt(t1,t2)

This method is used to calculate the 1st order derivative of kernel function given two one dimensional real inputs.

Public fields

k_par

vector(of length n_hy) containing the hyper-parameter of kernel. n_hy is the length of kernel hyper parameters.

Methods

Public methods

• Kernel$new()

• Kernel$greet()

• Kernel$kern()

• Kernel$dkd_kpar()

• Kernel$dkdt()

• Kernel$clone()

---

Method new()

Usage

Kernel$new(k_par = NULL)

---

Method greet()

Usage

Kernel$greet()

---

Method kern()

Usage

Kernel$kern(t1, t2)

---

Method dkd_kpar()

Usage

Kernel$dkd_kpar(t1, t2)

---

Method dkdt()

Usage

Kernel$dkdt(t1, t2)

---

Method clone()

The objects of this class are cloneable with this method.

Usage

Kernel$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Author(s)

Mu Niu, [e-mail]

KGode documentation built on Aug. 19, 2022, 5:08 p.m.