Hi users,
I ran into this error with a analyis using PRED routine.
0MINIMIZATION TERMINATED
DUE TO MAX. NO. OF FUNCTION EVALUATIONS EXCEEDED
NO. OF FUNCTION EVALUATIONS USED: 10093
NO. OF SIG. DIGITS UNREPORTABLE
I did find in previous discussion, people suggested to do by adding this under
$PK
" FIRST
" COMMON /PRCOMG/ IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" INTEGER IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" IMAX=100000
but, when I added above below $PRED, it didn't work.
Any suggestion? Thanks.
help on a minimization problem
Hi Ethan,
You have reached the maximum number of function evaluation (9999 I
believe). Two solutions:
1- you create a copy of your control stream and update the initial
estimates with the 'final' estimates of your first run
2- you create a copy of your control stream and use an $MSFI statement
in case you would have use the MSFO option in your first run.
Then, run the copied control stream. Hopefully, your run will converge
before reached the maximum number of function evaluation.
HIH
Sebastien
Ethan Wu wrote:
Hi users,
I ran into this error with a analyis using PRED routine.
0MINIMIZATION TERMINATED
DUE TO MAX. NO. OF FUNCTION EVALUATIONS EXCEEDED
NO. OF FUNCTION EVALUATIONS USED: 10093
NO. OF SIG. DIGITS UNREPORTABLE
I did find in previous discussion, people suggested to do by
adding this under $PK
" FIRST
" COMMON /PRCOMG/ IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" INTEGER IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" IMAX=100000
but, when I added above below $PRED, it didn't work.
Any suggestion? Thanks.
Ethan,
There are (at least) two kinds of evaluations/iterations you need to be aware of when using NONMEM.
1. The number of function evaluations which is a way of limiting the total work that NONMEM tries to make during estimation. It is determined by the estimation option MAXEVAL.
2. The number of iterations used by the differential equation solver when trying to make a step from one data record to the next.
The problem you report is with the number of function evaluations. In NM7 you can specify a number larger than 10000 to ask NONMEM to do more work.
The verbatim code trick you tried works with NMVI but not with NM7 to increase the number of differential equation solver iterations. Does anybody know how to get NM7 to try harder to solve differential equations?
Nick
Ethan Wu wrote:
> Hi users,
> I ran into this error with a analyis using PRED routine.
> 0MINIMIZATION TERMINATED
> DUE TO MAX. NO. OF FUNCTION EVALUATIONS EXCEEDED
> NO. OF FUNCTION EVALUATIONS USED: 10093
> NO. OF SIG. DIGITS UNREPORTABLE
>
> I did find in previous discussion, people suggested to do by adding this under $PK
>
> " FIRST
> " COMMON /PRCOMG/ IDUM1,IDUM2,IMAX,IDUM4,IDUM5
> " INTEGER IDUM1,IDUM2,IMAX,IDUM4,IDUM5
> " IMAX=100000
> but, when I added above below $PRED, it didn't work.
> Any suggestion? Thanks.
--
Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
tel:+64(9)923-6730 fax:+64(9)373-7090 mobile:+64(21)46 23 53
email: [email protected]
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford
Ethan and Nick,
To change the maximum number of calls during an integration interval for
NM7 try placing the following at the beginning of $PK.
"FIRST
" USE PRCOM_INT, ONLY: IMAX
and somewhere after all MU-referencing specify the IMAX value.
For example,
" IMAX=3000000
Beginning with version VI the default value is 1000000.
To change the default value, redefine MAXFCN in SIZES and recompile.
Tom
Quoted reply history
________________________________
From: [email protected] [mailto:[email protected]]
On Behalf Of Nick Holford
Sent: Friday, June 18, 2010 4:29 PM
To: [email protected]
Subject: Re: [NMusers] help on a minimization problem
Ethan,
There are (at least) two kinds of evaluations/iterations you need to be
aware of when using NONMEM.
1. The number of function evaluations which is a way of limiting the
total work that NONMEM tries to make during estimation. It is determined
by the estimation option MAXEVAL.
2. The number of iterations used by the differential equation solver
when trying to make a step from one data record to the next.
The problem you report is with the number of function evaluations. In
NM7 you can specify a number larger than 10000 to ask NONMEM to do more
work.
The verbatim code trick you tried works with NMVI but not with NM7 to
increase the number of differential equation solver iterations. Does
anybody know how to get NM7 to try harder to solve differential
equations?
Nick
Ethan Wu wrote:
Hi users,
I ran into this error with a analyis using PRED routine.
0MINIMIZATION TERMINATED
DUE TO MAX. NO. OF FUNCTION EVALUATIONS EXCEEDED
NO. OF FUNCTION EVALUATIONS USED: 10093
NO. OF SIG. DIGITS UNREPORTABLE
I did find in previous discussion, people suggested to do by
adding this under $PK
" FIRST
" COMMON /PRCOMG/ IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" INTEGER IDUM1,IDUM2,IMAX,IDUM4,IDUM5
" IMAX=100000
but, when I added above below $PRED, it didn't work.
Any suggestion? Thanks.
--
Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
tel:+64(9)923-6730 fax:+64(9)373-7090 mobile:+64(21)46 23 53
email: [email protected]
http://www.fmhs.auckland.ac.nz/sms/pharmacology/holford