RE: Can we increase maximum number of function evaluations?

From: Nick Holford n.holford@auckland.ac.nz Subject: RE: [NMusers] Can we increase maximum number of function evaluations? Date: Mon, 28 Aug 2006 10:08:46 +1200 Hi, Pavel is rather vague about the problem he is having and why he is interested in changing function evaluations. It seems there are two issues here: First, in NONMEM speak the number of function evaluations usually means the number of objective function evaluations. It is controlled by the $ESTIMATION option MAXEVALS. The maximum possible is MAXEVALS=9999. This can be used to limit the amount of work that NONMEM does when searching for parameters. Typically one uses either MAXEVALS=9999 or MAXEVALS=0. I dont have much use for other values. MAXEVALS=n2 Maximum allowable number of evaluations of the objective function during the Estimation Step. Default: a generous number. (Each evaluation of the objective function requires one pass through the data set.) This is also referred to as a "function evalua- tion.") MAXEVAL=0 omits the Estimation Step; it is useful with POSTHOC. MAXEVALS=-1 may be specified when a $MSFI record is present. It requests that NONMEM re-use the value from the pre- vious run, and is the default with $MSFI. Second, Mark has shown some verbatim code for increasing the number of differential equation evaluations used by the differential equation solver. This has little to do with the job that NONMEM is doing i.e. estimating the model parameters, but has to do with the work required to solve the differential equations. Mark gives an example for increasing the rather small number of DE evaluations that is supplied by default. The online help indicates the verbatim option FIRST should start immediately after the ". FIRST verbatim code Verbatim lines which must be positioned immediately after the declarations which are part of the normal subroutine header, and prior to the FIRST executable statement of the subroutine, must precede the first line of abbreviated code and must start with the line "FIRST. I am not sure if it makes any difference to do it the way that Mark wrote it. However, it is essential to have at least one blank between " and COMMON. If not then NM-TRAN will treat this record as a Fortran comment statement. Nick -- Nick Holford, Dept Pharmacology & Clinical Pharmacology University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand email:n.holford@auckland.ac.nz tel:+64(9)373-7599x86730 fax:373-7556 http://www.health.auckland.ac.nz/pharmacology/staff/nholford/