release version 0.1.0

DESCRIPTION

@@ -1,22 +1,22 @@
 Package: EPBO
 Title: Bayesian Optimization via Exact Penalty
-Version: 0.0.11
-Date: 2023-11-20
+Version: 0.1.0
+Date: 2023-12-14
 Authors@R: 
   person(given = "Jiangyan", 
          family = "Zhao", 
          email = "[email protected]", 
          role = c("aut", "cre"))
-Description: Wrapper routines for blackbox optimization under mixed equality and inequality 
-  constraints via an exact penalty.
+Description: Wrapper routines for blackbox optimization under mixed (equality and/or inequality) 
+  constraints via exact penalty.
 License: AGPL (>= 3)
 URL: https://github.com/Jiangyan-Zhao/EPBO
 BugReports: https://github.com/Jiangyan-Zhao/EPBO/issues
 Encoding: UTF-8
 LazyData: true
 Roxygen: list(markdown = TRUE)
 RoxygenNote: 7.2.3
-Imports: laGP, tgp, DiceKriging, matrixStats, mvtnorm
+Imports: laGP, tgp
 Depends: R (>= 2.14)
 Suggests: 
     stats,

NAMESPACE

@@ -1,38 +1,18 @@
 # Generated by roxygen2: do not edit by hand
 
-export(AF_AE)
-export(AF_ConVar)
-export(AF_EY)
-export(AF_EY_Kernel)
-export(AF_LCB)
-export(AF_OOSS)
-export(AF_ScaledEI)
-export(AF_ScaledEI_Kernel)
-export(AF_ScaledEI_MC)
-export(AF_TS)
-export(AF_UEI)
-export(AF_UEI_Kernel)
-export(AF_UEI_MC)
-export(AF_VEI)
-export(bilog)
-export(normalize)
 export(optim.AE)
 export(optim.BM)
 export(optim.EP)
-export(optim.EP.MC)
-export(optim.EP.kernel)
-export(optim.EP4HD)
-export(unbilog)
-export(unnormalize)
-import(DiceKriging)
 import(laGP)
-import(matrixStats)
 import(tgp)
+importFrom(graphics,image)
+importFrom(graphics,par)
+importFrom(graphics,points)
 importFrom(stats,dnorm)
+importFrom(stats,median)
 importFrom(stats,optim)
 importFrom(stats,pnorm)
 importFrom(stats,quantile)
-importFrom(stats,rnorm)
 importFrom(stats,runif)
 importFrom(stats,sd)
 importFrom(utils,tail)

R/AE.R

@@ -46,7 +46,7 @@
 #' \item{C }{ \code{matrix} giving the value of the constraint function for the input under consideration at each trial}
 #' \item{X }{ \code{matrix} giving the input values at which the blackbox function was evaluated }
 #' 
-#'  @seealso \code{\link[laGP]{optim.auglag}}, \code{\link[laGP]{optim.efi}}, \code{\link[EPBO]{optim.BM}}, \code{\link[EPBO]{optim.EP}}
+#' @seealso \code{\link[laGP]{optim.auglag}}, \code{\link[laGP]{optim.efi}}, \code{\link[EPBO]{optim.BM}}, \code{\link[EPBO]{optim.EP}}
 #'  
 #' @author Jiangyan Zhao \email{[email protected]}
 #' 
@@ -68,10 +68,9 @@
 #'
 #' @examples 
 #' # search space
-#' B = rbind(c(-2.25, 2.5), c(-2.25, 1.75))
+#' B = rbind(c(-2.25, 2.5), c(-2.5, 1.75))
 #' # Objective and constraint function for use 
-#' MTP = function(x, B=rbind(c(-2.25, 2.5), c(-2.5, 1.75))){
-#'   x = pmin(pmax(B[,1], x), B[,2])
+#' MTP = function(x){
 #'   f = -(cos((x[1]-0.1)*x[2]))^2 - x[1]*sin(3*x[1]+x[2])
 #'   t = atan2(x[1], x[2])
 #'   c = x[1]^2 + x[2]^2 -((2*cos(t)-1/2*cos(2*t)-1/4*cos(3*t)-1/8*cos(4*t))^2) - ((2*sin(t))^2)
@@ -84,14 +83,13 @@
 #' AE$xbest
 
 
-optim.AE = function(blackbox, B, 
-                    start=10, end=100, 
-                    Xstart=NULL, urate=10, 
-                    ncandf=function(k) { k }, 
-                    alpha1=1, alpha2=5, omega=2/3,
-                    dg.start=c(0.1,1e-6), ab=c(3/2,8), 
-                    dlim=sqrt(ncol(B))*c(1/100,10), 
-                    verb=2, ...){
+optim.AE = function(
+    blackbox, B, start=10, end=100, 
+    Xstart=NULL, urate=10, ncandf=function(k) { k }, 
+    alpha1=1, alpha2=5, omega=2/3,
+    dg.start=c(0.1,1e-6), ab=c(3/2,8), dlim=sqrt(ncol(B))*c(1/100,10), 
+    verb=2, ...)
+{
   ## check start
   if(start >= end) stop("must have start < end")
   ## get initial design

R/AF_AE.R

@@ -1,26 +1,21 @@
-#' @title Asymmetric Entropy
+#' @title Asymmetric entropy acquisition function
 #' 
-#' @description Asymmetric Entropy
+#' @description The asymmetric entropy (AE) acquisition function of the AE method
 #' 
-#' @param x description
+#' @param x a vector containing a single candidate point; or a \code{matrix} with 
+#' multiple candidate points
+#' @param fgpi the GP surrogate model of the objective function
+#' @param fnorm the maximum of the objective 
+#' @param Cgpi the GP surrogate models of the constraints
+#' @param Cnorm the maxima of the constraints
+#' @param fmin the best objective value obtained so far
+#' @param alpha1 a specified weight for the EI part
+#' @param alpha2 a specified weight for the AE part
+#' @param omega a mode location parameter
 #' 
-#' @param fgpi description
+#' @returns The AE value(s) at \code{x}. 
 #' 
-#' @param fnorm description
-#' 
-#' @param Cgpi description
-#' 
-#' @param Cnorm description
-#' 
-#' @param fmin description
-#' 
-#' @param alpha1 description
-#' 
-#' @param alpha2 description
-#' 
-#' @param omega description
-#' 
-#' @returns AF 
+#' @seealso \code{\link[EPBO]{AF_ScaledEI}}, \code{\link[EPBO]{AF_EY}}, \code{\link[EPBO]{AF_OOSS}}
 #' 
 #' @author Jiangyan Zhao \email{[email protected]}
 #' 
@@ -31,15 +26,7 @@
 #' @importFrom stats dnorm 
 #' @importFrom stats pnorm 
 #' @importFrom stats quantile
-#' 
-#' 
-#' 
-#' @export
-#' 
-#' @examples 
-#' B = rbind(c(0, 1), c(0, 1)) 
-#' 
-#' 
+
 
 AF_AE = function(x, fgpi, fnorm, Cgpi, Cnorm, fmin,
                  alpha1=1, alpha2=5, omega=2/3)

R/AF_EY.R

@@ -1,40 +1,28 @@
 #' @title Predictive mean acquisition function
 #' 
-#' @description Predictive mean
-#' 
-#' @param x description
-#' 
-#' @param fgpi description
-#' 
-#' @param fmean description
-#' 
-#' @param fsd description
-#' 
-#' @param Cgpi description
-#' 
-#' @param rho description
-#' 
+#' @description The predictive mean acquisition function of the EPBO method
+#' 
+#' @param x a vector containing a single candidate point; or a \code{matrix} with 
+#' multiple candidate points
+#' @param fgpi the GP surrogate model of the objective function
+#' @param fmean the mean of the objective value
+#' @param fsd the standard deviation of the objective value
+#' @param Cgpi the GP surrogate models of the constraints
+#' @param rho the penalty parameters
 #' @param equal an optional vector containing zeros and ones, whose length equals the number of
 #' constraints, specifying which should be treated as equality constraints (\code{1}) and 
 #' which as inequality (\code{0}) 
 #' 
+#' @returns The Predictive Mean at \code{x}. 
 #' 
-#' @returns AF 
+#' @seealso \code{\link[EPBO]{AF_ScaledEI}}, \code{\link[EPBO]{AF_OOSS}}, \code{\link[EPBO]{AF_AE}}
 #' 
 #' @author Jiangyan Zhao \email{[email protected]}
 #' 
-#' 
 #' @import laGP
 #' @importFrom stats dnorm 
 #' @importFrom stats pnorm 
-#' 
-#' 
-#' @export
-#' 
-#' @examples 
-#' B = rbind(c(0, 1), c(0, 1)) 
-#' 
-#' 
+
 
 AF_EY = function(x, fgpi, fmean, fsd, Cgpi, rho, equal)
 {