model.class: A Reference Class to generates differents model objects

In mathieucarmassi/CaliCo: Code Calibration in a Bayesian Framework

Description

See the function model which produces an instance of this class This class comes with a set of methods, some of them being useful for the user: See the documentation for model... Other methods should not be called as they are designed to be used during the calibration process.

Fields should not be changed or manipulated by the user as they are updated internally during the estimation process.

Usage

model.class

Format

An object of class R6ClassGenerator of length 24.

Fields

code

a function which takes in entry X and theta

X

the matrix of the forced variables

Yexp

the experimental output

n

the number of experiments

d

the number of forced variables

binf

the lower bound of the parameters for the DOE

bsup

the upper bound of the parameters for the DOE

opt.gp

a list of parameter for the surrogate (default NULL)

• type type of the chosen kernel (value by default "matern5_2") from km function

• DOE design of experiments for the surrogate (default value NULL)

opt.emul

a list of parameter to establish the DOE (default NULL)

• p the number of parameter in the model (default value 1)

• n.emul the number of points for constituting the Design Of Experiments (DOE) (default value 100)

• binf the lower bound of the parameter vector (default value 0)

• bsup the upper bound of the parameter vector (default value 1)

opt.sim

a list of parameter containing output of the code and corresponding DOE

• Ysim Output of the code

• DOEsim DOE corresponding to the output of the code

model

the model choice (see model for more specification).

opt.disc

a list of parameter for the discrepancy

• kernel.type the kernel chosen for the Gaussian process

mathieucarmassi/CaliCo documentation built on Aug. 14, 2019, 11:32 a.m.