Last updated on 2025-03-11 12:53:05 CET.
| Flavor | Version | Tinstall | Tcheck | Ttotal | Status | Flags |
|---|---|---|---|---|---|---|
| r-devel-linux-x86_64-debian-clang | 0.9-1 | 269.05 | 239.11 | 508.16 | OK | |
| r-devel-linux-x86_64-debian-gcc | 0.9-1 | 206.49 | 164.94 | 371.43 | OK | |
| r-devel-linux-x86_64-fedora-clang | 0.9-1 | 1021.50 | ERROR | |||
| r-devel-linux-x86_64-fedora-gcc | 0.9-1 | 975.68 | ERROR | |||
| r-devel-macos-arm64 | 0.9-1 | 186.00 | OK | |||
| r-devel-macos-x86_64 | 0.9-1 | 379.00 | OK | |||
| r-devel-windows-x86_64 | 0.9-1 | 370.00 | 270.00 | 640.00 | OK | |
| r-patched-linux-x86_64 | 0.9-1 | 305.39 | 214.38 | 519.77 | NOTE | |
| r-release-linux-x86_64 | 0.9-1 | 301.89 | 215.40 | 517.29 | NOTE | |
| r-release-macos-arm64 | 0.9-1 | 185.00 | NOTE | |||
| r-release-macos-x86_64 | 0.9-1 | 364.00 | NOTE | |||
| r-release-windows-x86_64 | 0.9-1 | 373.00 | 268.00 | 641.00 | NOTE | |
| r-oldrel-macos-arm64 | 0.9-1 | NOTE | ||||
| r-oldrel-macos-x86_64 | 0.9-1 | 331.00 | NOTE | |||
| r-oldrel-windows-x86_64 | 0.9-1 | 426.00 | 355.00 | 781.00 | NOTE |
Version: 0.9-1
Check: tests
Result: ERROR
Running ‘test-AllKrigingConcistency.R’ [8s/29s]
Running ‘test-KrigingCopy.R’ [5s/12s]
Running ‘test-KrigingFit.R’ [7s/16s]
Running ‘test-KrigingLeaveOneOut.R’ [7s/16s]
Running ‘test-KrigingLeaveOneOut_3d.R’
Running ‘test-KrigingLogLik.R’ [9s/13s]
Running ‘test-KrigingLogLikGradHess.R’ [26s/43s]
Running ‘test-KrigingMethods.R’ [9s/14s]
Running ‘test-KrigingPredict.R’ [30s/66s]
Running ‘test-KrigingSimulate.R’ [4s/11s]
Running ‘test-KrigingUpdate.R’
Running ‘test-KrigingUpdateSimulate.R’ [7s/16s]
Running ‘test-LinearAlgebra.R’ [5s/12s]
Running ‘test-NoiseKrigingFit.R’ [13s/22s]
Running ‘test-NoiseKrigingLogLik.R’ [8s/18s]
Running ‘test-NoiseKrigingMethods.R’ [7s/16s]
Running ‘test-NoiseKrigingPredict.R’ [19s/48s]
Running ‘test-NoiseKrigingSimulate.R’ [11s/25s]
Running ‘test-NoiseKrigingUpdate.R’
Running ‘test-NoiseKrigingUpdateSimulate.R’ [8s/16s]
Running ‘test-NuggetKrigingFit.R’ [14s/24s]
Running ‘test-NuggetKrigingLogLik.R’ [15s/26s]
Running ‘test-NuggetKrigingLogMargPost.R’ [23s/42s]
Running ‘test-NuggetKrigingMethods.R’ [7s/16s]
Running ‘test-NuggetKrigingPredict.R’ [20s/47s]
Running ‘test-NuggetKrigingSimulate.R’ [5s/12s]
Running ‘test-NuggetKrigingUpdate.R’
Running ‘test-NuggetKrigingUpdateSimulate.R’ [7s/17s]
Running ‘test-RobustGaSP-Nugget.R’ [5s/11s]
Running ‘test-RobustGaSP.R’ [5s/13s]
Running ‘test-RobustGaSPtrendlinear.R’ [5s/13s]
Running ‘test-RobustGaSPvsKrigingLMP.R’ [5s/11s]
Running ‘test-RobustGaSPvsNuggetKrigingLMP.R’ [5s/15s]
Running ‘test-SaveLoad.R’ [5s/11s]
Running ‘test-asDiceKriging.R’ [28s/72s]
Running ‘test-estimnone.R’ [5s/11s]
Running ‘test-normalize.R’ [8s/17s]
Running ‘test-rlibkriging-demo.R’
Running ‘test-unstableLL.R’ [4s/11s]
Running the tests in ‘tests/test-RobustGaSP.R’ failed.
Complete output:
> library(testthat)
> Sys.setenv('OMP_THREAD_LIMIT'=2)
> library(rlibkriging)
Attaching package: 'rlibkriging'
The following objects are masked from 'package:base':
load, save
>
> ##library(rlibkriging, lib.loc="bindings/R/Rlibs")
> ##library(testthat)
>
> library(RobustGaSP)
#########
##
## Robust Gaussian Stochastic Process, RobustGaSP Package
## Copyright (C) 2016-2025 Mengyang Gu, Jesus Palomo and James O. Berger
#########
Attaching package: 'RobustGaSP'
The following object is masked from 'package:rlibkriging':
simulate
The following object is masked from 'package:stats':
simulate
>
> context("RobustGaSP / Fit: 1D")
>
> f = function(x) 1-1/2*(sin(12*x)/(1+x)+2*cos(7*x)*x^5+0.7)
> #plot(f)
> n <- 5
> set.seed(123)
> X <- as.matrix(runif(n))
> y = f(X)
> #points(X,y)
> k = RobustGaSP::rgasp(design=X,response=y)
The upper bounds of the range parameters are 184.9743
The initial values of range parameters are 3.699485
Start of the optimization 1 :
The number of iterations is 30
The value of the marginal posterior function is 2.497978
Optimized range parameters are 0.1921691
Optimized nugget parameter is 0
Convergence: TRUE
The initial values of range parameters are 0.05223118
Start of the optimization 2 :
The number of iterations is 30
The value of the marginal posterior function is 1.035387
Optimized range parameters are 0.05296527
Optimized nugget parameter is 0
Convergence: TRUE
> #library(rlibkriging)
> r <- Kriging(y, X,
+ kernel="matern5_2",
+ regmodel = "constant", normalize = FALSE,
+ optim = "BFGS",
+ objective = "LMP")
OMP: Warning #96: Cannot form a team with 24 threads, using 2 instead.
OMP: Hint Consider unsetting KMP_DEVICE_THREAD_LIMIT (KMP_ALL_THREADS), KMP_TEAMS_THREAD_LIMIT, and OMP_THREAD_LIMIT (if any are set).
> # m = as.list(r)
>
> # Check lmp function
>
> lmp_rgasp = function(X, model=k) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+ #y=-logMargPostFun(r,matrix(unlist(x),ncol=2))$logMargPost
+ y=RobustGaSP:::neg_log_marginal_post_approx_ref(param=(x),nugget=0, nugget.est=model@nugget.est,
+ R0=model@R0,X=model@X, zero_mean=model@zero_mean,output=model@output,
+ CL=model@CL,
+ a=0.2,
+ b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2]),
+ kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha
+ )
+ y})}
> lmp_rgasp(1)
[1] -1.901254
>
> plot(lmp_rgasp,xlim=c(0.01,6))
> abline(v=(log(k@beta_hat)))
>
> lmp_lk = function(X) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+ y=-logMargPostFun(r,matrix(unlist(exp(-(x))),ncol=1))$logMargPost
+ y})}
> lmp_lk(1)
[1] -1.901254
>
> lines(seq(0.1,6,,5),lmp_lk(seq(0.1,6,,5)),col='red')
> abline(v=(log(1/as.list(r)$theta)),col='red')
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / Fit: 1D / rgasp/lmp is the same that lk/lmp one"),
+ expect_equal(lmp_rgasp(1),lmp_lk(1),tol = precision))
Test passed 🥳
> test_that(desc=paste0("RobustGaSP / Fit: 1D / fitted theta is the same that RobustGaSP one"),
+ expect_equal(as.list(r)$theta[1],1/k@beta_hat,tol = precision))
Test passed 🥳
>
>
>
> dlmp_rgasp = function(X, model=k) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+
+ # print(RobustGaSP:::log_marginal_lik_deriv(param=(x),nugget=0,nugget_est=model@nugget.est,
+ # R0=model@R0,X=model@X, zero_mean=model@zero_mean,
+ # output=model@output,
+ # kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha))
+ #
+ # print(RobustGaSP:::log_approx_ref_prior_deriv(param=(x),nugget=0, nugget_est=model@nugget.est,
+ # CL=model@CL,
+ # a=0.2,
+ # b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2])))
+
+
+ #y=-logMargPostFun(r,matrix(unlist(x),ncol=2))$logMargPost
+ y=RobustGaSP:::neg_log_marginal_post_approx_ref_deriv(param=(x),nugget=0, nugget.est=model@nugget.est,
+ R0=model@R0,X=model@X, zero_mean=model@zero_mean,output=model@output,
+ CL=model@CL,
+ a=0.2,
+ b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2]),
+ kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha
+ )
+ y})}
> dlmp_rgasp(1)
[1] -1.703845
>
> dlmp_lk = function(X) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ apply(X,1,
+ function(x) {
+ y=-logMargPostFun(r,matrix(unlist(exp(-(x))),ncol=1),TRUE)$logMargPostGrad
+ y})}
> -exp(-1)*dlmp_lk(1)
[1] -1.703845
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / Fit: 1D / rgasp/lmp deriv is the same that lk/lmp deriv"),
+ expect_equal(dlmp_rgasp(1),-exp(-1)*dlmp_lk(1),tol = precision))
Test passed 🎉
>
>
> # Check predict
>
> ntest <- 10
> Xtest <- seq(0,1,,ntest)
> Ytest_rgasp <- predict(k,matrix(Xtest,ncol=1))
> Ytest_libK <- predict(r,Xtest)
>
> plot(f)
> points(X,y)
> lines(Xtest,Ytest_rgasp$mean,col='blue')
> polygon(c(Xtest,rev(Xtest)),
+ c(Ytest_rgasp$mean+2*Ytest_rgasp$sd,rev(Ytest_rgasp$mean-2*Ytest_rgasp$sd)),
+ col=rgb(0,0,1,0.1), border=NA)
>
> lines(Xtest,Ytest_libK$mean,col='red')
> polygon(c(Xtest,rev(Xtest)),
+ c(Ytest_libK$mean+2*Ytest_libK$stdev,rev(Ytest_libK$mean-2*Ytest_libK$stdev)),
+ col=rgb(1,0,0,0.1), border=NA)
>
> precision <- 1e-3
> test_that(desc=paste0("pred mean is the same that RobustGaSP one"),
+ expect_equal(predict(r,0.7)$mean[1],predict(k,matrix(0.7))$mean,tol = precision))
Test passed 😸
> test_that(desc=paste0("pred sd is the same that RobustGaSP one"),
+ expect_equal(predict(r,0.7)$stdev[1],predict(k,matrix(0.7))$sd,tol = precision))
Test passed 🥇
>
>
> ## RobustGaSP examples
>
> #---------------------------------------
> # a 1 dimensional example
> #---------------------------------------
> context("RobustGaSP / 1 dimensional example")
>
>
> input=10*seq(0,1,1/14)
> output<-higdon.1.data(input)
> #the following code fit a GaSP with zero mean by setting zero.mean="Yes"
> model<- rgasp(design = input, response = output, zero.mean="No")
The upper bounds of the range parameters are 670.0756
The initial values of range parameters are 13.40151
Start of the optimization 1 :
The number of iterations is 30
The value of the marginal posterior function is -10.48964
Optimized range parameters are 13.2106
Optimized nugget parameter is 0
Convergence: TRUE
The initial values of range parameters are 0.08888889
Start of the optimization 2 :
The number of iterations is 30
The value of the marginal posterior function is -15.24592
Optimized range parameters are 0.1706386
Optimized nugget parameter is 0
Convergence: TRUE
> model
Call:
rgasp(design = input, response = output, zero.mean = "No")
Mean parameters: 2.174187e-10
Variance parameter: 4249.587
Range parameters: 13.2106
Noise parameter: 0
>
> testing_input = as.matrix(seq(0,10,1/100))
> model.predict<-predict(model,testing_input)
> names(model.predict)
[1] "mean" "lower95" "upper95" "sd"
>
> #########plot predictive distribution
> testing_output=higdon.1.data(testing_input)
> plot(testing_input,model.predict$mean,type='l',col='blue',
+ xlab='input',ylab='output')
> polygon( c(testing_input,rev(testing_input)),c(model.predict$lower95,
+ rev(model.predict$upper95)),col = "grey80", border = FALSE)
> lines(testing_input, testing_output)
> lines(testing_input,model.predict$mean,type='l',col='blue')
> lines(input, output,type='p')
>
> ## mean square erros
> mean((model.predict$mean-testing_output)^2)
[1] 4.63608e-05
>
> model_libK = Kriging(matrix(output,ncol=1), matrix(input,ncol=1),
+ kernel="matern5_2",
+ regmodel = "constant", normalize = FALSE,
+ optim = "BFGS",
+ objective = "LMP", parameters = NULL)
>
> lines(testing_input,predict(model_libK,testing_input)$mean,type='l',col='red')
> polygon(
+ c(testing_input,rev(testing_input)),
+ c(
+ predict(model_libK,testing_input)$mean+2*predict(model_libK,testing_input)$stdev,
+ rev(predict(model_libK,testing_input)$mean-2*predict(model_libK,testing_input)$stdev)),
+ col = rgb(1,0,0,0.1), border = FALSE)
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / 1 dimensional example / pred mean is the same that RobustGaSP one"),
+ expect_equal(predict(model_libK,0.7)$mean[1],predict(model,matrix(0.7))$mean,tol = precision))
── Failure: RobustGaSP / 1 dimensional example / pred mean is the same that RobustGaSP one ──
predict(model_libK, 0.7)$mean[1] not equal to predict(model, matrix(0.7))$mean.
1/1 mismatches
[1] 0.621 - 0.623 == -0.00162
Error:
! Test failed
Backtrace:
▆
1. ├─testthat::test_that(...)
2. │ └─withr (local) `<fn>`()
3. └─reporter$stop_if_needed()
4. └─rlang::abort("Test failed", call = NULL)
Execution halted
Flavor: r-devel-linux-x86_64-fedora-clang
Version: 0.9-1
Check: tests
Result: ERROR
Running ‘test-AllKrigingConcistency.R’ [8s/10s]
Running ‘test-KrigingCopy.R’
Running ‘test-KrigingFit.R’
Running ‘test-KrigingLeaveOneOut.R’
Running ‘test-KrigingLeaveOneOut_3d.R’
Running ‘test-KrigingLogLik.R’
Running ‘test-KrigingLogLikGradHess.R’ [27s/28s]
Running ‘test-KrigingMethods.R’
Running ‘test-KrigingPredict.R’ [25s/29s]
Running ‘test-KrigingSimulate.R’
Running ‘test-KrigingUpdate.R’
Running ‘test-KrigingUpdateSimulate.R’
Running ‘test-LinearAlgebra.R’
Running ‘test-NoiseKrigingFit.R’ [12s/14s]
Running ‘test-NoiseKrigingLogLik.R’
Running ‘test-NoiseKrigingMethods.R’
Running ‘test-NoiseKrigingPredict.R’ [17s/21s]
Running ‘test-NoiseKrigingSimulate.R’ [10s/12s]
Running ‘test-NoiseKrigingUpdate.R’
Running ‘test-NoiseKrigingUpdateSimulate.R’
Running ‘test-NuggetKrigingFit.R’ [15s/16s]
Running ‘test-NuggetKrigingLogLik.R’ [11s/13s]
Running ‘test-NuggetKrigingLogMargPost.R’ [18s/21s]
Running ‘test-NuggetKrigingMethods.R’
Running ‘test-NuggetKrigingPredict.R’ [17s/22s]
Running ‘test-NuggetKrigingSimulate.R’
Running ‘test-NuggetKrigingUpdate.R’
Running ‘test-NuggetKrigingUpdateSimulate.R’
Running ‘test-RobustGaSP-Nugget.R’
Running ‘test-RobustGaSP.R’
Running ‘test-RobustGaSPtrendlinear.R’
Running ‘test-RobustGaSPvsKrigingLMP.R’
Running ‘test-RobustGaSPvsNuggetKrigingLMP.R’
Running ‘test-SaveLoad.R’
Running ‘test-asDiceKriging.R’ [22s/26s]
Running ‘test-estimnone.R’
Running ‘test-normalize.R’
Running ‘test-rlibkriging-demo.R’
Running ‘test-unstableLL.R’
Running the tests in ‘tests/test-RobustGaSP.R’ failed.
Complete output:
> library(testthat)
> Sys.setenv('OMP_THREAD_LIMIT'=2)
> library(rlibkriging)
Attaching package: 'rlibkriging'
The following objects are masked from 'package:base':
load, save
>
> ##library(rlibkriging, lib.loc="bindings/R/Rlibs")
> ##library(testthat)
>
> library(RobustGaSP)
#########
##
## Robust Gaussian Stochastic Process, RobustGaSP Package
## Copyright (C) 2016-2025 Mengyang Gu, Jesus Palomo and James O. Berger
#########
Attaching package: 'RobustGaSP'
The following object is masked from 'package:rlibkriging':
simulate
The following object is masked from 'package:stats':
simulate
>
> context("RobustGaSP / Fit: 1D")
>
> f = function(x) 1-1/2*(sin(12*x)/(1+x)+2*cos(7*x)*x^5+0.7)
> #plot(f)
> n <- 5
> set.seed(123)
> X <- as.matrix(runif(n))
> y = f(X)
> #points(X,y)
> k = RobustGaSP::rgasp(design=X,response=y)
The upper bounds of the range parameters are 184.9743
The initial values of range parameters are 3.699485
Start of the optimization 1 :
The number of iterations is 30
The value of the marginal posterior function is 2.497978
Optimized range parameters are 0.1921691
Optimized nugget parameter is 0
Convergence: TRUE
The initial values of range parameters are 0.05223118
Start of the optimization 2 :
The number of iterations is 30
The value of the marginal posterior function is 1.035387
Optimized range parameters are 0.05296527
Optimized nugget parameter is 0
Convergence: TRUE
> #library(rlibkriging)
> r <- Kriging(y, X,
+ kernel="matern5_2",
+ regmodel = "constant", normalize = FALSE,
+ optim = "BFGS",
+ objective = "LMP")
> # m = as.list(r)
>
> # Check lmp function
>
> lmp_rgasp = function(X, model=k) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+ #y=-logMargPostFun(r,matrix(unlist(x),ncol=2))$logMargPost
+ y=RobustGaSP:::neg_log_marginal_post_approx_ref(param=(x),nugget=0, nugget.est=model@nugget.est,
+ R0=model@R0,X=model@X, zero_mean=model@zero_mean,output=model@output,
+ CL=model@CL,
+ a=0.2,
+ b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2]),
+ kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha
+ )
+ y})}
> lmp_rgasp(1)
[1] -1.901254
>
> plot(lmp_rgasp,xlim=c(0.01,6))
> abline(v=(log(k@beta_hat)))
>
> lmp_lk = function(X) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+ y=-logMargPostFun(r,matrix(unlist(exp(-(x))),ncol=1))$logMargPost
+ y})}
> lmp_lk(1)
[1] -1.901254
>
> lines(seq(0.1,6,,5),lmp_lk(seq(0.1,6,,5)),col='red')
> abline(v=(log(1/as.list(r)$theta)),col='red')
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / Fit: 1D / rgasp/lmp is the same that lk/lmp one"),
+ expect_equal(lmp_rgasp(1),lmp_lk(1),tol = precision))
Test passed 🥳
> test_that(desc=paste0("RobustGaSP / Fit: 1D / fitted theta is the same that RobustGaSP one"),
+ expect_equal(as.list(r)$theta[1],1/k@beta_hat,tol = precision))
Test passed 🥳
>
>
>
> dlmp_rgasp = function(X, model=k) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ # print(dim(X));
+ apply(X,1,
+ function(x) {
+
+ # print(RobustGaSP:::log_marginal_lik_deriv(param=(x),nugget=0,nugget_est=model@nugget.est,
+ # R0=model@R0,X=model@X, zero_mean=model@zero_mean,
+ # output=model@output,
+ # kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha))
+ #
+ # print(RobustGaSP:::log_approx_ref_prior_deriv(param=(x),nugget=0, nugget_est=model@nugget.est,
+ # CL=model@CL,
+ # a=0.2,
+ # b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2])))
+
+
+ #y=-logMargPostFun(r,matrix(unlist(x),ncol=2))$logMargPost
+ y=RobustGaSP:::neg_log_marginal_post_approx_ref_deriv(param=(x),nugget=0, nugget.est=model@nugget.est,
+ R0=model@R0,X=model@X, zero_mean=model@zero_mean,output=model@output,
+ CL=model@CL,
+ a=0.2,
+ b=1/(length(model@output))^{1/dim(as.matrix(model@input))[2]}*(0.2+dim(as.matrix(model@input))[2]),
+ kernel_type=rep(as.integer(3),ncol(X)),alpha=model@alpha
+ )
+ y})}
> dlmp_rgasp(1)
[1] -1.703845
>
> dlmp_lk = function(X) {if (!is.matrix(X)) X = matrix(X,ncol=1);
+ apply(X,1,
+ function(x) {
+ y=-logMargPostFun(r,matrix(unlist(exp(-(x))),ncol=1),TRUE)$logMargPostGrad
+ y})}
> -exp(-1)*dlmp_lk(1)
[1] -1.703845
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / Fit: 1D / rgasp/lmp deriv is the same that lk/lmp deriv"),
+ expect_equal(dlmp_rgasp(1),-exp(-1)*dlmp_lk(1),tol = precision))
Test passed 🎉
>
>
> # Check predict
>
> ntest <- 10
> Xtest <- seq(0,1,,ntest)
> Ytest_rgasp <- predict(k,matrix(Xtest,ncol=1))
> Ytest_libK <- predict(r,Xtest)
>
> plot(f)
> points(X,y)
> lines(Xtest,Ytest_rgasp$mean,col='blue')
> polygon(c(Xtest,rev(Xtest)),
+ c(Ytest_rgasp$mean+2*Ytest_rgasp$sd,rev(Ytest_rgasp$mean-2*Ytest_rgasp$sd)),
+ col=rgb(0,0,1,0.1), border=NA)
>
> lines(Xtest,Ytest_libK$mean,col='red')
> polygon(c(Xtest,rev(Xtest)),
+ c(Ytest_libK$mean+2*Ytest_libK$stdev,rev(Ytest_libK$mean-2*Ytest_libK$stdev)),
+ col=rgb(1,0,0,0.1), border=NA)
>
> precision <- 1e-3
> test_that(desc=paste0("pred mean is the same that RobustGaSP one"),
+ expect_equal(predict(r,0.7)$mean[1],predict(k,matrix(0.7))$mean,tol = precision))
Test passed 😸
> test_that(desc=paste0("pred sd is the same that RobustGaSP one"),
+ expect_equal(predict(r,0.7)$stdev[1],predict(k,matrix(0.7))$sd,tol = precision))
Test passed 🥇
>
>
> ## RobustGaSP examples
>
> #---------------------------------------
> # a 1 dimensional example
> #---------------------------------------
> context("RobustGaSP / 1 dimensional example")
>
>
> input=10*seq(0,1,1/14)
> output<-higdon.1.data(input)
> #the following code fit a GaSP with zero mean by setting zero.mean="Yes"
> model<- rgasp(design = input, response = output, zero.mean="No")
The upper bounds of the range parameters are 670.0756
The initial values of range parameters are 13.40151
Start of the optimization 1 :
The number of iterations is 30
The value of the marginal posterior function is -10.48964
Optimized range parameters are 13.2106
Optimized nugget parameter is 0
Convergence: TRUE
The initial values of range parameters are 0.08888889
Start of the optimization 2 :
The number of iterations is 30
The value of the marginal posterior function is -15.24592
Optimized range parameters are 0.1706386
Optimized nugget parameter is 0
Convergence: TRUE
> model
Call:
rgasp(design = input, response = output, zero.mean = "No")
Mean parameters: 2.174187e-10
Variance parameter: 4249.587
Range parameters: 13.2106
Noise parameter: 0
>
> testing_input = as.matrix(seq(0,10,1/100))
> model.predict<-predict(model,testing_input)
> names(model.predict)
[1] "mean" "lower95" "upper95" "sd"
>
> #########plot predictive distribution
> testing_output=higdon.1.data(testing_input)
> plot(testing_input,model.predict$mean,type='l',col='blue',
+ xlab='input',ylab='output')
> polygon( c(testing_input,rev(testing_input)),c(model.predict$lower95,
+ rev(model.predict$upper95)),col = "grey80", border = FALSE)
> lines(testing_input, testing_output)
> lines(testing_input,model.predict$mean,type='l',col='blue')
> lines(input, output,type='p')
>
> ## mean square erros
> mean((model.predict$mean-testing_output)^2)
[1] 4.63608e-05
>
> model_libK = Kriging(matrix(output,ncol=1), matrix(input,ncol=1),
+ kernel="matern5_2",
+ regmodel = "constant", normalize = FALSE,
+ optim = "BFGS",
+ objective = "LMP", parameters = NULL)
>
> lines(testing_input,predict(model_libK,testing_input)$mean,type='l',col='red')
> polygon(
+ c(testing_input,rev(testing_input)),
+ c(
+ predict(model_libK,testing_input)$mean+2*predict(model_libK,testing_input)$stdev,
+ rev(predict(model_libK,testing_input)$mean-2*predict(model_libK,testing_input)$stdev)),
+ col = rgb(1,0,0,0.1), border = FALSE)
>
> precision <- 1e-3
> test_that(desc=paste0("RobustGaSP / 1 dimensional example / pred mean is the same that RobustGaSP one"),
+ expect_equal(predict(model_libK,0.7)$mean[1],predict(model,matrix(0.7))$mean,tol = precision))
── Failure: RobustGaSP / 1 dimensional example / pred mean is the same that RobustGaSP one ──
predict(model_libK, 0.7)$mean[1] not equal to predict(model, matrix(0.7))$mean.
1/1 mismatches
[1] 0.621 - 0.623 == -0.00162
Error:
! Test failed
Backtrace:
▆
1. ├─testthat::test_that(...)
2. │ └─withr (local) `<fn>`()
3. └─reporter$stop_if_needed()
4. └─rlang::abort("Test failed", call = NULL)
Execution halted
Flavor: r-devel-linux-x86_64-fedora-gcc
Version: 0.9-1
Check: for GNU extensions in Makefiles
Result: NOTE
GNU make is a SystemRequirements.
Flavors: r-patched-linux-x86_64, r-release-linux-x86_64, r-release-macos-arm64, r-release-macos-x86_64, r-release-windows-x86_64, r-oldrel-macos-arm64, r-oldrel-macos-x86_64, r-oldrel-windows-x86_64
Version: 0.9-1
Check: installed package size
Result: NOTE
installed size is 71.3Mb
sub-directories of 1Mb or more:
include 7.3Mb
lib 37.3Mb
libs 26.2Mb
Flavors: r-release-macos-arm64, r-release-macos-x86_64, r-release-windows-x86_64, r-oldrel-macos-arm64, r-oldrel-macos-x86_64, r-oldrel-windows-x86_64
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.