R: Function summary for Multi-Fidelity Cokriging models
summary.MuFicokm
R Documentation
Function summary for Multi-Fidelity Cokriging models
Description
Provide a summary of a multi-fidelity cokriging model. In particular, it provides the parameter estimations and the results of the
cross-validation procedure.
Usage
## S3 method for class 'MuFicokm'
summary(object, CrossValidation = FALSE, ...)
Arguments
object
an object of class S3 ("MuFicokm") provided by the function MuFicokm corresponding to the multi-fidelity cokriging model.
CrossValidation
a Boolean. If TRUE, a Leave-One-Out cross validation procedure is performed. For the LOO procedure, the responses are removed from all code levels and the trend, adjustment and variance parameters are re-estimated after each removed observation.
...
no other argument for this method.
Details
"summary.MuFicokm" return the parameter estimations for each level and the result of the Leave-One-Out Cross-Validation
(RMSE=Root Mean Squared Error ; Std RMSE=Standardized RMSE ;
Q2=explained variance).
Value
A list with following items (see "MuFicokm"):
CovNames
a list of character strings giving the covariance structures used for the cokriging model. The element i of the list corresponds to the covariance structure of the Gaussian process δ_i(x) with δ_1(x) = Z_1(x). (see "MuFicokm")
Cov.val
a list of vectors giving the values of the hyper-parameters of the cokriging model. The element i of the list corresponds to the hyper-parameters of the Gaussian process δ_i(x) with δ_1(x) = Z_1(x). (see "MuFicokm")
Var.val
a list of numerics giving the values of the variance parameters of the cokriging model. The element i of the list corresponds to the variance of the Gaussian process δ_i(x) with δ_1(x) = Z_1(x). (see "MuFicokm")
Rho.val
a list of vectors giving the values of the trends γ_i of the adjustment parameters ρ_i of the cokriging model. The element i of the list corresponds to the adjustment parameter between Z_i and δ_i(x). (see "MuFicokm")
Trend.val
a list of vectors giving the values of the trend parameters of the Gaussian processes δ_i(x) and Z_1(x).
Author(s)
Loic Le Gratiet
Examples
#--- test functions (see [Le GRATIET, L. 2012])
Funcf <- function(x){return(0.5*(6*x-2)^2*sin(12*x-4)+sin(10*cos(5*x)))}
Funcc <- function(x){return((6*x-2)^2*sin(12*x-4)+10*(x-0.5)-5)}
#--- Data
Dc <- seq(0,1,0.1)
indDf <- c(1,3,7,11)
DNest <- NestedDesign(Dc, nlevel=2 , indices = list(indDf) )
zc <- Funcc(DNest$PX)
Df <- ExtractNestDesign(DNest,2)
zf <- Funcf(Df)
#--- Multi-fidelity cokriging creation without parameter estimations
mymodel <- MuFicokm(
formula = list(~1,~1),
MuFidesign = DNest,
response = list(zc,zf),
nlevel = 2)
sum <- summary(object = mymodel, CrossValidation = TRUE)
names(sum)
#--- Saving parameters
#--covariance parameters
sum$Cov.Val
#--variance parameters
sum$Var.Val
#--trend parameters
sum$Trend.Val
#-- adjustment parameters
sum$Rho.Val
Results
R version 3.3.1 (2016-06-21) -- "Bug in Your Hair"
Copyright (C) 2016 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)
R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
> library(MuFiCokriging)
Loading required package: DiceKriging
> png(filename="/home/ddbj/snapshot/RGM3/R_CC/result/MuFiCokriging/summary.MuFicokm.Rd_%03d_medium.png", width=480, height=480)
> ### Name: summary.MuFicokm
> ### Title: Function summary for Multi-Fidelity Cokriging models
> ### Aliases: summary.MuFicokm
>
> ### ** Examples
>
> #--- test functions (see [Le GRATIET, L. 2012])
> Funcf <- function(x){return(0.5*(6*x-2)^2*sin(12*x-4)+sin(10*cos(5*x)))}
> Funcc <- function(x){return((6*x-2)^2*sin(12*x-4)+10*(x-0.5)-5)}
> #--- Data
> Dc <- seq(0,1,0.1)
> indDf <- c(1,3,7,11)
> DNest <- NestedDesign(Dc, nlevel=2 , indices = list(indDf) )
> zc <- Funcc(DNest$PX)
> Df <- ExtractNestDesign(DNest,2)
> zf <- Funcf(Df)
> #--- Multi-fidelity cokriging creation without parameter estimations
> mymodel <- MuFicokm(
+ formula = list(~1,~1),
+ MuFidesign = DNest,
+ response = list(zc,zf),
+ nlevel = 2)
optimisation start
------------------
* estimation method : MLE
* optimisation method : BFGS
* analytical gradient : used
* trend model : ~1
* covariance model :
- type : matern5_2
- nugget : NO
- parameters lower bounds : 1e-10
- parameters upper bounds : 2
- best initial criterion value(s) : -30.12058
N = 1, M = 5 machine precision = 2.22045e-16
At X0, 0 variables are exactly at the bounds
At iterate 0 f= 30.121 |proj g|= 0.33827
At iterate 1 f = 30.079 |proj g|= 0.32754
At iterate 2 f = 30.036 |proj g|= 0.8379
At iterate 3 f = 30.034 |proj g|= 0.10589
At iterate 4 f = 30.033 |proj g|= 0.0027327
At iterate 5 f = 30.033 |proj g|= 9.3487e-06
iterations 5
function evaluations 7
segments explored during Cauchy searches 5
BFGS updates skipped 0
active bounds at final generalized Cauchy point 0
norm of the final projected gradient 9.34872e-06
final function value 30.0335
F = 30.0335
final value 30.033468
converged
optimisation start
------------------
* estimation method : MLE
* optimisation method : BFGS
* analytical gradient : used
* trend model : ~1
* covariance model :
- type : matern5_2
- nugget : NO
- parameters lower bounds : 1e-10
- parameters upper bounds : 2
- best initial criterion value(s) : -3.813998
N = 1, M = 5 machine precision = 2.22045e-16
At X0, 0 variables are exactly at the bounds
At iterate 0 f= 3.814 |proj g|= 3.5182e-21
Derivative >= 0, backtracking line search impossible.final value 3.813998
stopped after 0 iterations
>
> sum <- summary(object = mymodel, CrossValidation = TRUE)
Level 1: parameter estimation
Covariance type: matern5_2
Variance estimation: 142.5591
Trend estimation: 1.856445
Correlation length estimation: 0.3014263
Level 2 : parameter estimation
Covariance type: matern5_2
Variance estimation: 0.3942074
Trend estimation: 2.473242
Adjustment estimation: 0.3601547
Correlation length estimation: 0.007034777
Leave One Cross Validation
RMSE:0.9018081
Std RMSE:0.7447738
Q2:0.9299403
> names(sum)
[1] "CovNames" "Cov.Val" "Var.Val" "Rho.Val" "Trend.Val"
> #--- Saving parameters
> #--covariance parameters
> sum$Cov.Val
[[1]]
[1] 0.3014263
[[2]]
[1] 0.007034777
> #--variance parameters
> sum$Var.Val
[[1]]
[1] 142.5591
[[2]]
[1] 0.3942074
> #--trend parameters
> sum$Trend.Val
[[1]]
[1] 1.856445
[[2]]
[1] 2.473242
> #-- adjustment parameters
> sum$Rho.Val
[[1]]
[1] 0.3601547
>
>
>
>
>
> dev.off()
null device
1
>