changes to make package build work

DESCRIPTION

@@ -4,12 +4,8 @@ Title: Population (and individual) optimal Experimental Design
 Version: 0.1.1
 Date: 2014-05-27
 Depends: ggplot2
-Imports: MASS, mvtnorm
+Imports: MASS, mvtnorm, nlme
 Suggests: testthat, Hmisc
-Author: Andrew C. Hooker, Sebastian Ueckert (MATLAB version), 
-  Marco Foracchia (O-Matrix version) and Joakim Nyberg (MATLAB version) 
-  with contributions from Eric Stroemberg (MATLAB version).
-Maintainer: Andrew C. Hooker <[email protected]>
 Authors@R: c(person("Andrew C.","Hooker",
                     email="[email protected]",
                     role=c("aut","cre","trl","cph")),

NAMESPACE

@@ -1,4 +1,5 @@
 import(ggplot2)
 importFrom(MASS, write.matrix)
 importFrom(mvtnorm, rmvnorm)
+importFrom(nlme, fdHess)
 exportPattern("^[[:alpha:]]+")

NEWS

@@ -7,6 +7,7 @@ PopED 0.1.1
 
 * New functionality to compute the ED OFV using the Laplace approximation.
   This can be orders of magnitude faster than the standard MC integration approach.
+  See '?ed_laplace_ofv' and '?evaluate.e.ofv.fim' 
 
 * Added a general function to compute the FIM and OFV(FIM) for all avaialbale methods in PopED.
   See '?calc_ofv_and_fim'.

R/LEDoptim.R

@@ -15,9 +15,11 @@
 #' @inheritParams RS_opt_gen
 #' @inheritParams create.poped.database
 #' @inheritParams Doptim
+#' @inheritParams calc_ofv_and_fim
 #' @param fim_init The initial value of the FIM. If set to zero then it is computed.
 #' @param ofv_init The inital OFV. If set to zero then it is computed.
 #' @param trflag Should the optimization be output to the screen and to a file?
+#' @param use_RS should the fucntion use a random search algorithm?
 #' 
 #' @family Optimize
 #' 
@@ -203,7 +205,7 @@ LEDoptim <- function(poped.db,
       fprintf('Starting BGFS minimization with OFV of %g \n', best_dmf)
       returnArgs <- bfgsb_min('ed_laplace_ofv',
                               list(model_switch,aa,axt,poped.db$groupsize,ni,
-                                   xtopt,xopt,aopt,bpop,d,poped.db$sigma,docc_full,poped.db,
+                                   xtopt,xopt,aopt,bpopdescr,ddescr,poped.db$sigma,poped.db$param.pt.val$docc,poped.db,
                                    return_gradient=T,
                                    optxt=optxt, opta=optx, x=x),
                               x_k,lb,ub,options) 

R/RS_opt_gen.R

@@ -12,12 +12,19 @@
 #' @inheritParams evaluate.fim
 #' @inheritParams Doptim
 #' @inheritParams create.poped.database
+#' @inheritParams calc_ofv_and_fim
 #' @param cfaxt First step factor for sample times 
 #' @param opt_xt Should the sample times be optimized?
 #' @param opt_a Should the continuous design variables be optimized?
 #' @param opt_x Should the discrete design variables be optimized?
 #' @param approx_type Approximation method for model, 0=FO, 1=FOCE, 2=FOCEI, 3=FOI.
 #' @param iter The number of iterations entered into the \code{blockheader_2} function.
+#' @param header_flag Should the header text be printed out?
+#' @param footer_flag Should the footer text be printed out?
+#' @param out_file Which file should the output be directed to?  A string, a file handle using 
+#'        \code{\link{file}} or \code{""} will output to the screen.
+#' @param compute_inv should the inverse of the FIM be used to compute expected RSE values?  Often not needed
+#'        except for diagnostic purposes.
 #' @param ... arguments passed to \code{\link{evaluate.fim}} and \code{\link{ofv_fim}}.
 #' 
 #' 

R/blockfinal.R

@@ -8,6 +8,7 @@
 #' @inheritParams create.poped.database
 #' @inheritParams blockexp
 #' @inheritParams blockheader
+#' @inheritParams RS_opt_gen
 #' @param fmf_init Initial FIM.
 #' @param dmf_init Initial OFV.
 #' @param param_cvs_init The inital design parameter RSE values.

R/blockheader.R

@@ -7,11 +7,13 @@
 #' @inheritParams blockexp
 #' @inheritParams Doptim
 #' @inheritParams create.poped.database
+#' @inheritParams RS_opt_gen
 #' @param name The name used for the output file. Combined with \code{name_header} and \code{iter}. 
 #' If \code{""} then output is to the screen.
 #' @param iter The last number in the name printed to the output file, combined with \code{name}.
 #' @param name_header The initial portion of the file name.
 #' @param file_path The path to where the file should be created.
+#' @param header_flag Should the header text be printed out?
 #' @param ... Additional arguments passed to further functions.
 #' 
 #' @family Helper

R/calc_ofv_and_fim.R

@@ -8,6 +8,8 @@
 #' @inheritParams evaluate.fim
 #' @inheritParams Doptim
 #' @inheritParams create.poped.database
+#' @param ofv The current ofv.  If other than zero then this values is simply returned unchanged.
+#' @param fim The current FIM. If other than zero then this values is simply returned unchanged.
 #' @param use_laplace Should the Laplace method be used in calculating the expectation of the OFV?  
 #' @param laplace.fim Should an E(FIM) be calculated when computing the Laplace approximated E(OFV).  Typically
 #' the FIM does not need to be computed and, if desired,  this calculation

R/ed_laplace_ofv.R

@@ -20,7 +20,10 @@
 #' @param optxt If sampling times are optimized
 #' @param opta If continuous design variables are optimized
 #' @param aopto the continuous design variables
-#'   
+#' @param method If 0 then use an optimization routine translated from poped code written in MATLAB to
+#'        optimize the parameters in the Laplace approximation.  If 1 then use \code{\link{optim}} to compute both
+#'        k and the hessian of k (see Dodds et al, JPP, 2005 for more information). If 2 then use \code{\link{fdHess}}
+#'        to compute the hessian.
 #' @param return_gradient Should the gradient be returned.
 #'   
 #' @return The FIM and the hessian of the FIM.
@@ -224,7 +227,7 @@ ed_laplace_ofv <- function(model_switch,groupsize,ni,xtopto,xopto,aopto,
 
     ## Different hessian and gradient calculation
     if(method==2){ 
-      k_vals <- nlme::fdHess(output$par,
+      k_vals <- fdHess(output$par,
                              function(x) calc_k(x,model_switch,groupsize,ni,xtopto,xopto,
                                                 aopto,bpopdescr,ddescr,covd,sigma,docc,poped.db,Engine,
                                                 return_gradient=F)[["k"]]) 

R/log_prior_pdf.R

@@ -4,7 +4,10 @@
 #' @inheritParams evaluate.fim
 #' @inheritParams create.poped.database
 #' @inheritParams Doptim
+#' @param alpha  A parameter vector.
+#' @param return_hessian Should the hessian be returned?
 #' 
+#' @example tests/testthat/examples_fcn_doc/warfarin_optimize.R
 #' @example tests/testthat/examples_fcn_doc/examples_log_prior_pdf.R
 
 log_prior_pdf <- function(alpha, bpopdescr, ddescr,return_gradient=F,return_hessian=F){

R/poped_optimize.R

@@ -11,6 +11,7 @@
 #' @inheritParams Doptim
 #' @inheritParams create.poped.database
 #' @inheritParams Dtrace
+#' @inheritParams calc_ofv_and_fim
 #' @param ... arguments passed to other functions. See \code{\link{Doptim}}.
 #' 
 #' 

man/LEDoptim.Rd

@@ -1,7 +1,7 @@
 % Generated by roxygen2 (4.0.1): do not edit by hand
 \name{LEDoptim}
 \alias{LEDoptim}
-\title{Laplace approximated ED-optimization function}
+\title{Optimization function for D-family, E-family and Laplace approximated ED designs}
 \usage{
 LEDoptim(poped.db, model_switch = NULL, ni = NULL, xt = NULL, x = NULL,
   a = NULL, bpopdescr = NULL, ddescr = NULL, maxxt = NULL,
@@ -19,9 +19,7 @@ LEDoptim(poped.db, model_switch = NULL, ni = NULL, xt = NULL, x = NULL,
 
 \item{trflag}{Should the optimization be output to the screen and to a file?}
 
-\item{ls_step_size}{Number of grid points in the line search}
-
-\item{iter_tot}{Number of iterations of full Random search and full Stochastic Gradient if line search is not used.}
+\item{use_RS}{should the fucntion use a random search algorithm?}
 
 \item{opt_xt}{Should the sample times be optimized?}
 
@@ -31,6 +29,13 @@ LEDoptim(poped.db, model_switch = NULL, ni = NULL, xt = NULL, x = NULL,
 
 \item{...}{arguments passed to \code{\link{evaluate.fim}} and \code{\link{ofv_fim}}.}
 
+\item{header_flag}{Should the header text be printed out?}
+
+\item{footer_flag}{Should the footer text be printed out?}
+
+\item{out_file}{Which file should the output be directed to?  A string, a file handle using
+\code{\link{file}} or \code{""} will output to the screen.}
+
 \item{model_switch}{Matrix defining which response a certain sampling time belongs to.}
 
 \item{ni}{Vector defining the number of samples for each group.}
@@ -77,9 +82,15 @@ D-family design (1) or ED-familty design (0) (with or without parameter uncertai
 }}
 
 \item{ddescr}{Matrix defining the diagnonals of the IIV (same logic as for the \code{bpopdescr}).}
+
+\item{use_laplace}{Should the Laplace method be used in calculating the expectation of the OFV?}
+
+\item{laplace.fim}{Should an E(FIM) be calculated when computing the Laplace approximated E(OFV).  Typically
+the FIM does not need to be computed and, if desired,  this calculation
+is done usng the standard MC integration technique, so can be slow.}
 }
 \description{
-Optimize the Laplace approximated ED-objective function.
+Optimize the objective fucntion for D-family, E-family and Laplace approximated ED designs.
 Right now there is only one optimization algorithm used in this
 function
 \enumerate{
@@ -114,11 +125,19 @@ sfg <- function(x,a,bpop,b,bocc){
   return(parameters) 
 }
 
+# Adding 10\% log-normal Uncertainty to fixed effects (not Favail)
+bpop_vals <- c(CL=0.15, V=8, KA=1.0, Favail=1)
+bpop_vals_ed_ln <- cbind(ones(length(bpop_vals),1)*4, # log-normal distribution
+                         bpop_vals,
+                         ones(length(bpop_vals),1)*(bpop_vals*0.1)^2) # 10\% of bpop value
+bpop_vals_ed_ln["Favail",]  <- c(0,1,0)
+bpop_vals_ed_ln
+
 ## -- Define initial design  and design space
 poped.db <- create.poped.database(ff_file="ff.PK.1.comp.oral.sd.CL",
                                   fg_file="sfg",
                                   fError_file="feps.add.prop",
-                                  bpop=c(CL=0.15, V=8, KA=1.0, Favail=1), 
+                                  bpop=bpop_vals_ed_ln, 
                                   notfixed_bpop=c(1,1,1,0),
                                   d=c(CL=0.07, V=0.02, KA=0.6), 
                                   sigma=c(0.01,0.25),
@@ -129,78 +148,45 @@ poped.db <- create.poped.database(ff_file="ff.PK.1.comp.oral.sd.CL",
                                   a=70,
                                   mina=0,
                                   maxa=100)
-# warfarin optimize model
+# warfain ed model
 
 \dontrun{
 
-  ##############
-  # typically one will use poped_optimize 
-  # This then calls Doptim for continuous optimization problems
-  ##############
+  LEDoptim(poped.db) 
 
+  LEDoptim(poped.db,opt_xt=T,rsit=10) 
 
-  # RS+SG+LS optimization of sample times
-  # optimization with just a few iterations
-  # only to check that things are working
-  output <- poped_optimize(poped.db,opt_xt=T,
-                           rsit=5,sgit=5,ls_step_size=5)
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE) 
+
+  LEDoptim(poped.db,opt_xt=T,rsit=10,laplace.fim=TRUE) 
 
-  # RS+SG+LS optimization of sample times 
-  # (longer run time than above but more likely to reach a maximum)
-  output <- poped_optimize(poped.db,opt_xt=T)
-  get_rse(output$fmf,output$poped.db)
-  plot_model_prediction(output$poped.db)
+  LEDoptim(poped.db,opt_xt=T,rsit=10,use_laplace=FALSE) 
 
+  ## testing header and footer
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           out_file="foobar.txt") 
 
-  # Random search (just a few samples here)
-  rs.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,rsit=20,
-                              bUseRandomSearch= 1,
-                              bUseStochasticGradient = 0,
-                              bUseBFGSMinimizer = 0,
-                              bUseLineSearch = 0)
+  ff <- LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+                 trflag=FALSE) 
 
-  # line search, DOSE and sample time optimization
-  ls.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,
-                              bUseRandomSearch= 0,
-                              bUseStochasticGradient = 0,
-                              bUseBFGSMinimizer = 0,
-                              bUseLineSearch = 1,
-                              ls_step_size=10)
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           header_flag=FALSE) 
 
-  # Stochastic gradient search, DOSE and sample time optimization
-  sg.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1, 
-                              bUseRandomSearch= 0,
-                              bUseStochasticGradient = 1,
-                              bUseBFGSMinimizer = 0,
-                              bUseLineSearch = 0,
-                              sgit=20)
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           out_file="") 
 
-  # BFGS search, DOSE and sample time optimization
-  bfgs.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,
-                                bUseRandomSearch= 0,
-                                bUseStochasticGradient = 0,
-                                bUseBFGSMinimizer = 1,
-                                bUseLineSearch = 0)
-
-  ##############
-  # If you really want to you can use Doptim dirtectly
-  ##############
-  dsl <- downsizing_general_design(poped.db)
-  poped.db$optsw[2] <- 1  # sample time optimization
-  output <- Doptim(poped.db,dsl$ni, dsl$xt, dsl$model_switch, dsl$x, dsl$a, 
-         dsl$bpop, dsl$d, dsl$maxxt, dsl$minxt,dsl$maxa,dsl$mina) 
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           footer_flag=FALSE) 
 
-}
-
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           footer_flag=FALSE, header_flag=FALSE) 
+
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           footer_flag=FALSE, header_flag=FALSE,out_file="foobar.txt") 
+
+  LEDoptim(poped.db,opt_xt=T,rsit=10,d_switch=TRUE,
+           footer_flag=FALSE, header_flag=FALSE,out_file="") 
 
-}
-\references{
-\enumerate{
-\item M. Foracchia, A.C. Hooker, P. Vicini and A. Ruggeri, "PopED, a software for optimal
-experimental design in population kinetics", Computer Methods and Programs in Biomedicine, 74, 2004.
-\item J. Nyberg, S. Ueckert, E.A. Stroemberg, S. Hennig, M.O. Karlsson and A.C. Hooker, "PopED: An extended,
-parallelized, nonlinear mixed effects models optimal design tool",
-Computer Methods and Programs in Biomedicine, 108, 2012.
 }
 }
 \seealso{

man/PopED-internal.Rd

@@ -46,7 +46,6 @@
 \alias{numel}
 \alias{new_ferror_file}
 \alias{min_function}
-\alias{log_prior_pdf}
 \alias{linspace}
 \alias{line_search_uc}
 \alias{line_search}

man/RS_opt_gen.Rd

@@ -28,6 +28,16 @@ RS_opt_gen(poped.db, ni = NULL, xt = NULL, model_switch = NULL,
 
 \item{iter}{The number of iterations entered into the \code{blockheader_2} function.}
 
+\item{header_flag}{Should the header text be printed out?}
+
+\item{footer_flag}{Should the footer text be printed out?}
+
+\item{out_file}{Which file should the output be directed to?  A string, a file handle using
+\code{\link{file}} or \code{""} will output to the screen.}
+
+\item{compute_inv}{should the inverse of the FIM be used to compute expected RSE values?  Often not needed
+except for diagnostic purposes.}
+
 \item{...}{arguments passed to \code{\link{evaluate.fim}} and \code{\link{ofv_fim}}.}
 
 \item{poped.db}{A PopED database.}
@@ -94,6 +104,12 @@ all a values are given the same max value}
 \item{d_switch}{\itemize{
 \item \bold{******START OF CRITERION SPECIFICATION OPTIONS**********}}
 D-family design (1) or ED-familty design (0) (with or without parameter uncertainty)}
+
+\item{use_laplace}{Should the Laplace method be used in calculating the expectation of the OFV?}
+
+\item{laplace.fim}{Should an E(FIM) be calculated when computing the Laplace approximated E(OFV).  Typically
+the FIM does not need to be computed and, if desired,  this calculation
+is done usng the standard MC integration technique, so can be slow.}
 }
 \description{
 Optimize the objective function using an adaptive random search algorithm.

man/blockfinal.Rd

@@ -62,6 +62,14 @@ can also just supply the parameter values as a \code{c()}.}
 can also just supply the diagnonal parameter values (variances) as a \code{c()}.}
 
 \item{fn}{The file handle to write to.}
+
+\item{compute_inv}{should the inverse of the FIM be used to compute expected RSE values?  Often not needed
+except for diagnostic purposes.}
+
+\item{out_file}{Which file should the output be directed to?  A string, a file handle using
+\code{\link{file}} or \code{""} will output to the screen.}
+
+\item{footer_flag}{Should the footer text be printed out?}
 }
 \description{
 Create some output to the screen and a text file that summarizes the problem you solved.
@@ -109,7 +117,7 @@ poped.db <- create.poped.database(ff_file="ff.PK.1.comp.oral.sd.CL",
 
 
 FIM <- evaluate.fim(poped.db) 
- <- det(FIM)
+dmf <- det(FIM)
 
 
 blockfinal(fn="",fmf=FIM,