version update for CRAN

Author: mingdeyu

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: dgpsi
 Type: Package
 Title: Interface to 'dgpsi' for Deep and Linked Gaussian Process Emulations 
-Version: 2.1.6-9000
+Version: 2.2.0
 Authors@R: c(
   person('Deyu', 'Ming', role = c('aut', 'cre', 'cph'),
          email = 'deyu.ming.16@ucl.ac.uk'),
@@ -12,7 +12,8 @@ Maintainer: Deyu Ming <deyu.ming.16@ucl.ac.uk>
 Description: Interface to the 'python' package 'dgpsi' for Gaussian process, deep Gaussian process, 
              and linked Gaussian process emulations of computer models and systems of computer
              models. The implementations follow Ming & Guillas (2021) <doi:10.1137/20M1323771> and 
-             Ming, Williamson, & Guillas (2022) <doi:10.1080/00401706.2022.2124311>. To get started with the package, 
+             Ming, Williamson, & Guillas (2023) <doi:10.1080/00401706.2022.2124311> and 
+             Ming & Williamson (2023) <arXiv:2306.01212>. To get started with the package, 
              see <https://mingdeyu.github.io/dgpsi-R/>.
 License: MIT + file LICENSE
 URL: https://github.com/mingdeyu/dgpsi-R, https://mingdeyu.github.io/dgpsi-R/

NEWS.md

@@ -1,4 +1,4 @@
-# dgpsi 2.1.6-9000 (development version)
+# dgpsi 2.2.0
 
 - The efficiency and speed of imputations involved in the training and predictions of DGP emulators are significantly improved (achieving roughly 3x faster training and imputations) by utilizing blocked Gibbs sampling that imputes latent variables layer-wise rather than node-wise. The blocked Gibbs sampling is now the default method for DGP emulator inference and can be changed back to the old node-wise approach by setting `blocked_gibbs = FALSE` in `dgp()`.
 - One can now optimize GP components that are contained in the same layer of a DGP emulator in parallel during the DGP emulator training, using multiple cores by setting the new argument `cores` in `dgp()`. This option is useful and can accelerate the training speed when the input dimension is moderately large (in which case there is a large number of GP components to be optimized) and the optimization of GP components is computationally expensive, e.g., when `share = FALSE` in which case input dimensions to individual GP components have different lengthscales.

R/initi_py.R

@@ -37,13 +37,13 @@ init_py <- function(py_ver = NULL, dgpsi_ver = NULL, reinstall = FALSE, uninstal
   if ( is.null(py_ver) ) py_ver <- '3.9.13'
   if ( is.null(dgpsi_ver) ) {
     ##For devel version
-    dgpsi_ver <- c('cython>=0.29.30', 'dill>=0.3.2', 'jupyter>=1.0.0', 'matplotlib-base>=3.2.1', 'numba >=0.51.2',
-                   'numpy >=1.18.2', 'pathos >=0.2.9', 'psutil >=5.8.0', 'pybind11 >=2.10.0', 'pythran >=0.11.0',
-                   'scikit-build >=0.15.0', 'scikit-learn >=0.22.0', 'scipy >=1.4.1', 'tqdm >=4.50.2', 'tabulate >=0.8.7')
-    env_name <- 'dgp_si_R_2_1_6_9000'
+    #dgpsi_ver <- c('cython>=0.29.30', 'dill>=0.3.2', 'jupyter>=1.0.0', 'matplotlib-base>=3.2.1', 'numba >=0.51.2',
+    #               'numpy >=1.18.2', 'pathos >=0.2.9', 'psutil >=5.8.0', 'pybind11 >=2.10.0', 'pythran >=0.11.0',
+    #               'scikit-build >=0.15.0', 'scikit-learn >=0.22.0', 'scipy >=1.4.1', 'tqdm >=4.50.2', 'tabulate >=0.8.7')
+    #env_name <- 'dgp_si_R_2_1_6_9000'
     ##For release version
-    #dgpsi_ver <- 'dgpsi==2.1.6'
-    #env_name <- 'dgp_si_R_2_1_6'
+    dgpsi_ver <- 'dgpsi==2.2.0'
+    env_name <- 'dgp_si_R_2_2_0'
   } else {
     env_name <- paste('dgp_si_R_', gsub(".", "_", dgpsi_ver,fixed=TRUE), sep = "")
     dgpsi_ver <- paste('dgpsi==', dgpsi_ver, sep = "")

R/utils.R

@@ -214,6 +214,7 @@ write <- function(object, pkl_file) {
 #' @md
 #' @export
 set_seed <- function(seed) {
+  seed <- as.integer(seed)
   set.seed(seed)
   reticulate::py_set_seed(seed, disable_hash_randomization = TRUE)
   pkg.env$dgpsi$nb_seed(seed)

README.md

@@ -5,6 +5,7 @@
   [![DOC](https://img.shields.io/badge/DOC-release-brightgreen)](https://mingdeyu.github.io/dgpsi-R/)
   [![REF](https://img.shields.io/badge/REF-Linked%20GP-informational)](https://doi.org/10.1137/20M1323771)
   [![REF](https://img.shields.io/badge/REF-Deep%20GP-informational)](https://doi.org/10.1080/00401706.2022.2124311)
+  [![REF](https://img.shields.io/badge/REF-Linked%20DGP-informational)](http://arxiv.org/abs/2306.01212)
   [![python](https://img.shields.io/badge/Python-dgpsi%20v2.1.6-informational)](https://github.com/mingdeyu/DGP)
   ![CRAN_License](https://img.shields.io/cran/l/dgpsi?color=green)
 
@@ -77,6 +78,8 @@ to install and activate the required Python environment. That's it, the package
 > ```
 
 ## References
-> [Ming, D., Williamson, D., and Guillas, S. (2022) Deep Gaussian process emulation using stochastic imputation. <i>Technometrics</i>. 0(0), 1-12.](https://doi.org/10.1080/00401706.2022.2124311)
+> [Ming, D. and Williamson, D. (2023) Linked deep Gaussian process emulation for model networks. arXiv:2306.01212](http://arxiv.org/abs/2306.01212)
+
+> [Ming, D., Williamson, D., and Guillas, S. (2023) Deep Gaussian process emulation using stochastic imputation. <i>Technometrics</i>. 65(2), 150-161.](https://doi.org/10.1080/00401706.2022.2124311)
 
 > [Ming, D. and Guillas, S. (2021) Linked Gaussian process emulation for systems of computer models using Mat&eacute;rn kernels and adaptive design, <i>SIAM/ASA Journal on Uncertainty Quantification</i>. 9(4), 1615-1642.](https://doi.org/10.1137/20M1323771)

inst/CITATION

@@ -22,13 +22,26 @@ bibentry(bibtype = "Article",
                    as.person("Daniel Williamson"),
                    as.person("Serge Guillas")),
   journal      = "Technometrics",
-  year         = "2022",
-  volume       = "0",
-  number       = "0",
-  pages        = "1--12",
+  year         = "2023",
+  volume       = "65",
+  number       = "2",
+  pages        = "150--161",
 
   textVersion  =
-  paste("Ming, D., Williamson, D., and Guillas, S. (2022)",
+  paste("Ming, D., Williamson, D., and Guillas, S. (2023)",
         "Deep Gaussian process emulation using stochastic imputation,",
-        "Technometrics. (0)0, 1-12.")
+        "Technometrics. (65)2, 150-161.")
+)
+
+bibentry(bibtype = "Unpublished",
+  title        = "Linked deep Gaussian process emulation for model networks",
+  author       = c(as.person("Deyu Ming"),
+                   as.person("Daniel Williamson")),
+  note         = "arXiv:2306.01212",
+  year         = "2023",
+
+  textVersion  =
+  paste("Ming, D. and Williamson, D. (2023)",
+        "Linked deep Gaussian process emulation for model networks,",
+        "arXiv:2306.01212")
 )

vignettes/dgpsi.Rmd

@@ -16,7 +16,7 @@ knitr::opts_chunk$set(
 )
 ```
 
-`dgpsi` provides a flexible toolbox for Gaussian process (GP), deep Gaussian process (DGP) and linked (D)GP emulations. In this guide, we show how to use the package to emulate a step function with a three-layered DGP structure. There are other examples showing the functionality of the package in [`Articles`](https://mingdeyu.github.io/dgpsi-R/dev/articles/index.html) on the package website, and how to customize a DGP structure. A comprehensive reference of available functions is documented in [`Reference`](https://mingdeyu.github.io/dgpsi-R/dev/reference/index.html) section of the package website. 
+`dgpsi` provides a flexible toolbox for Gaussian process (GP), deep Gaussian process (DGP) and linked (D)GP emulations. In this guide, we show how to use the package to emulate a step function with a three-layered DGP structure. There are other examples showing the functionality of the package in [`Articles`](https://mingdeyu.github.io/dgpsi-R/articles/index.html) on the package website, and how to customize a DGP structure. A comprehensive reference of available functions is documented in [`Reference`](https://mingdeyu.github.io/dgpsi-R/reference/index.html) section of the package website. 
 
 ## Load the package
 

vignettes/seq_design.Rmd

@@ -262,7 +262,7 @@ It can be seen from the plot above that with static space-filling designs, the q
 
 ### See also
 
-See [`Sequential Design II`](https://mingdeyu.github.io/dgpsi-R/dev/articles/seq_design_2.html) for the sequential design of a bundle of DGP emulators with automatic terminations.
+See [`Sequential Design II`](https://mingdeyu.github.io/dgpsi-R/articles/seq_design_2.html) for the sequential design of a bundle of DGP emulators with automatic terminations.
 
 ### References
 

vignettes/seq_design_2.Rmd

@@ -324,4 +324,4 @@ It can be seen from the plot above that with the sequential design, emulators in
 
 ### See also
 
-See [`Sequential Design I`](https://mingdeyu.github.io/dgpsi-R/dev/articles/seq_design.html) for the sequential design of a DGP emulator on a 2D simulator.
+See [`Sequential Design I`](https://mingdeyu.github.io/dgpsi-R/articles/seq_design.html) for the sequential design of a DGP emulator on a 2D simulator.