Doptim | R Documentation |
Optimize the objective function. There are 4 different optimization algorithms used in this function
Adaptive random search.
See RS_opt
.
Stochastic gradient.
A Broyden Fletcher Goldfarb Shanno (BFGS) method for nonlinear minimization with box constraints.
A line search. See a_line_search
.
The
optimization algorithms run in series, taking as input the output from the
previous method. The stopping rule used is to test if the line search
algorithm fids a better optimum then its initial value. If so, then the chain
of algorithms is run again. If line search is not used then the argument
iter_tot
defines the number of times the chain of algorithms is run.
This function takes information from the PopED database supplied as an
argument. The PopED database supplies information about the the model,
parameters, design and methods to use. Some of the arguments coming from the
PopED database can be overwritten; if they are supplied then they are used
instead of the arguments from the PopED database.
Doptim(
poped.db,
ni,
xt,
model_switch,
x,
a,
bpopdescr,
ddescr,
maxxt,
minxt,
maxa,
mina,
fmf = 0,
dmf = 0,
trflag = TRUE,
bUseRandomSearch = poped.db$settings$bUseRandomSearch,
bUseStochasticGradient = poped.db$settings$bUseStochasticGradient,
bUseBFGSMinimizer = poped.db$settings$bUseBFGSMinimizer,
bUseLineSearch = poped.db$settings$bUseLineSearch,
sgit = poped.db$settings$sgit,
ls_step_size = poped.db$settings$ls_step_size,
BFGSConvergenceCriteriaMinStep = poped.db$settings$BFGSConvergenceCriteriaMinStep,
BFGSProjectedGradientTol = poped.db$settings$BFGSProjectedGradientTol,
BFGSTolerancef = poped.db$settings$BFGSTolerancef,
BFGSToleranceg = poped.db$settings$BFGSToleranceg,
BFGSTolerancex = poped.db$settings$BFGSTolerancex,
iter_tot = poped.db$settings$iNumSearchIterationsIfNotLineSearch,
iter_max = 10,
...
)
poped.db |
A PopED database. |
ni |
A vector of the number of samples in each group. |
xt |
A matrix of sample times. Each row is a vector of sample times for a group. |
model_switch |
A matrix that is the same size as xt, specifying which model each sample belongs to. |
x |
A matrix for the discrete design variables. Each row is a group. |
a |
A matrix of covariates. Each row is a group. |
bpopdescr |
Matrix defining the fixed effects, per row (row number = parameter_number) we should have:
|
ddescr |
Matrix defining the diagonals of the IIV (same logic as for
the |
maxxt |
Matrix or single value defining the maximum value for each xt sample. If a single value is supplied then all xt values are given the same maximum value. |
minxt |
Matrix or single value defining the minimum value for each xt sample. If a single value is supplied then all xt values are given the same minimum value |
maxa |
Vector defining the max value for each covariate. If a single value is supplied then all a values are given the same max value |
mina |
Vector defining the min value for each covariate. If a single value is supplied then all a values are given the same max value |
fmf |
The initial value of the FIM. If set to zero then it is computed. |
dmf |
The initial OFV. If set to zero then it is computed. |
trflag |
Should the optimization be output to the screen and to a file? |
bUseRandomSearch |
Use random search (1=TRUE, 0=FALSE) |
bUseStochasticGradient |
Use Stochastic Gradient search (1=TRUE, 0=FALSE) |
bUseBFGSMinimizer |
Use BFGS Minimizer (1=TRUE, 0=FALSE) |
bUseLineSearch |
Use Line search (1=TRUE, 0=FALSE) |
sgit |
Number of stochastic gradient iterations |
ls_step_size |
Number of grid points in the line search. |
BFGSConvergenceCriteriaMinStep |
BFGS Minimizer Convergence Criteria Minimum Step |
BFGSProjectedGradientTol |
BFGS Minimizer Convergence Criteria Normalized Projected Gradient Tolerance |
BFGSTolerancef |
BFGS Minimizer Line Search Tolerance f |
BFGSToleranceg |
BFGS Minimizer Line Search Tolerance g |
BFGSTolerancex |
BFGS Minimizer Line Search Tolerance x |
iter_tot |
Number of iterations to use if line search is not used. Must
be less than |
iter_max |
If line search is used then the algorithm tests if line
search (always run at the end of the optimization iteration) changes the
design in any way. If not, the algorithm stops. If yes, then a new
iteration is run unless |
... |
arguments passed to |
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.
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.
Other Optimize:
LEDoptim()
,
RS_opt()
,
a_line_search()
,
bfgsb_min()
,
calc_autofocus()
,
calc_ofv_and_grad()
,
mfea()
,
optim_ARS()
,
optim_LS()
,
poped_optim()
,
poped_optim_1()
,
poped_optim_2()
,
poped_optim_3()
,
poped_optimize()
library(PopED)
############# START #################
## Create PopED database
## (warfarin model for optimization)
#####################################
## Warfarin example from software comparison in:
## Nyberg et al., "Methods and software tools for design evaluation
## for population pharmacokinetics-pharmacodynamics studies",
## Br. J. Clin. Pharm., 2014.
## Optimization using an additive + proportional reidual error
## to avoid sample times at very low concentrations (time 0 or very late samples).
## find the parameters that are needed to define from the structural model
ff.PK.1.comp.oral.sd.CL
## -- parameter definition function
## -- names match parameters in function ff
sfg <- function(x,a,bpop,b,bocc){
parameters=c(CL=bpop[1]*exp(b[1]),
V=bpop[2]*exp(b[2]),
KA=bpop[3]*exp(b[3]),
Favail=bpop[4],
DOSE=a[1])
return(parameters)
}
## -- Define initial design and design space
poped.db <- create.poped.database(ff_fun=ff.PK.1.comp.oral.sd.CL,
fg_fun=sfg,
fError_fun=feps.add.prop,
bpop=c(CL=0.15, V=8, KA=1.0, Favail=1),
notfixed_bpop=c(1,1,1,0),
d=c(CL=0.07, V=0.02, KA=0.6),
sigma=c(prop=0.01,add=0.25),
groupsize=32,
xt=c( 0.5,1,2,6,24,36,72,120),
minxt=0.01,
maxxt=120,
a=c(DOSE=70),
mina=c(DOSE=0.01),
maxa=c(DOSE=100))
############# END ###################
## Create PopED database
## (warfarin model for optimization)
#####################################
## Not run:
##############
# typically one will use poped_optimize
# This then calls Doptim for continuous optimization problems
##############
# 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)
# 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$poped.db)
# Random search (just a few samples here)
rs.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,rsit=20,
bUseRandomSearch= 1,
bUseStochasticGradient = 0,
bUseBFGSMinimizer = 0,
bUseLineSearch = 0)
# line search, DOSE and sample time optimization
ls.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,
bUseRandomSearch= 0,
bUseStochasticGradient = 0,
bUseBFGSMinimizer = 0,
bUseLineSearch = 1,
ls_step_size=10)
# Stochastic gradient search, DOSE and sample time optimization
sg.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,
bUseRandomSearch= 0,
bUseStochasticGradient = 1,
bUseBFGSMinimizer = 0,
bUseLineSearch = 0,
sgit=20)
# BFGS search, DOSE and sample time optimization
bfgs.output <- poped_optimize(poped.db,opt_xt=1,opt_a=1,
bUseRandomSearch= 0,
bUseStochasticGradient = 0,
bUseBFGSMinimizer = 1,
bUseLineSearch = 0)
##############
# If you really want to you can use Doptim dirtectly
##############
dsl <- downsizing_general_design(poped.db)
poped.db$settings$optsw[2] <- 1 # sample time optimization
output <- Doptim(poped.db,dsl$ni, dsl$xt, dsl$model_switch, dsl$x, dsl$a,
dsl$bpop, dsl$d, dsl$maxxt, dsl$minxt,dsl$maxa,dsl$mina)
## End(Not run)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.