Dear NMUsers,
I am having trouble with PSN4.9, if anyone has a similar issue may I know
how to fix it?
I ran a model that was developed prior to the release of PSN4.9. In PSN4.9
the covariance step is aborted because S Matrix is unobtainable. I purged
4.9 and loaded 4.8, the model ran perfectly and matched the
previous results. Thank you very much in advance.
*Version 4.9*
#TERE:
Elapsed estimation time in seconds: 633.22
0S MATRIX UNOBTAINABLE
0COVARIANCE STEP ABORTED
Elapsed covariance time in seconds: 47.98
Elapsed postprocess time in seconds: 51.79
*Version 4.8*
#TERM:
0MINIMIZATION SUCCESSFUL
NO. OF FUNCTION EVALUATIONS USED: 783
NO. OF SIG. DIGITS IN FINAL EST.: 2.0
Regards,
Shankar Lanke
Problem with PSN4. 9 - COVARIANCE STEP ABORTED
Dear Shankar,
I think you will need to go deeper into the configuration and settings, as well
as to the directories where NONMEM was actually run, to answer this question.
Here are some things that you can consider:
Did you run both PsN versions using the same nonmem version?
If termination was not successful, was PsN (execute) set up to try again with
another set of initial estimates, and if it was: did you use the same seed when
running execute-4.8 and execute-4.9?
The control streams created by PsN (located in the NM_run1 directory): are they
identical? In this directory you will find e.g. psn.mod. If on a LINUX system
you can just run diff or ’sdiff -s’ to compare this file between the two runs
with execute
Did you change anything else in the configuration of PsN or NONMEM, e.g. the
compiler?
Best wishes
Jakob
Jakob Ribbing, Ph.D.
Senior Consultant, Pharmetheus AB
Cell/Mobile: +46 (0)70 514 33 77
[email protected]
www.pharmetheus.com
Phone, Office: +46 (0)18 513 328
Uppsala Science Park, Dag Hammarskjölds väg 36B
SE-752 37 Uppsala, Sweden
Quoted reply history
> On 3 May 2021, at 18:45, Shankar Lanke <[email protected]> wrote:
>
> Dear NMUsers,
> I am having trouble with PSN4.9, if anyone has a similar issue may I know how
> to fix it?
> I ran a model that was developed prior to the release of PSN4.9. In PSN4.9
> the covariance step is aborted because S Matrix is unobtainable. I purged 4.9
> and loaded 4.8, the model ran perfectly and matched the previous results.
> Thank you very much in advance.
>
> Version 4.9
> #TERE:
> Elapsed estimation time in seconds: 633.22
> 0S MATRIX UNOBTAINABLE
> 0COVARIANCE STEP ABORTED
> Elapsed covariance time in seconds: 47.98
> Elapsed postprocess time in seconds: 51.79
>
> Version 4.8
> #TERM:
> 0MINIMIZATION SUCCESSFUL
> NO. OF FUNCTION EVALUATIONS USED: 783
> NO. OF SIG. DIGITS IN FINAL EST.: 2.0
>
> Regards, <>
> Shankar Lanke
--
*This communication is confidential and is only intended for the use of the
individual or entity to which it is directed. It may contain information
that is privileged and exempt from disclosure under applicable law. If you
are not the intended recipient please notify us immediately. Please do not
copy it or disclose its contents to any other person.*
Dear Jakob,
1. Did you run both PsN versions using the same nonmem version? *Yes*
2. If termination was not successful, was PsN (execute) set up to try
again with another set of initial estimates *[Yes, tried with different
IEs, Fixed few, all combinations]*, and if it was: did you use the same
seed when running execute-4.8 and execute-4.9? *Yes*
3. ’sdiff -s’ to compare this file between the two runs with execute-
*Yes*
4. Did you change anything else in the configuration of PsN or NONMEM,
e.g. the compiler? *Computing system is located in Linux server, I dont
think if IT makes any changes to it without informing us, but I will check
with them. *
Thank you so much for your time and input.
Quoted reply history
On Tue, May 4, 2021 at 8:58 AM Jakob Ribbing <[email protected]>
wrote:
> Dear Shankar,
>
> I think you will need to go deeper into the configuration and settings, as
> well as to the directories where NONMEM was actually run, to answer this
> question.
>
> Here are some things that you can consider:
>
> - Did you run both PsN versions using the same nonmem version?
> - If termination was not successful, was PsN (execute) set up to try
> again with another set of initial estimates, and if it was: did you use the
> same seed when running execute-4.8 and execute-4.9?
> - The control streams created by PsN (located in the NM_run1
> directory): are they identical? In this directory you will find e.g.
> psn.mod. If on a LINUX system you can just run diff or ’sdiff -s’ to
> compare this file between the two runs with execute
> - Did you change anything else in the configuration of PsN or NONMEM,
> e.g. the compiler?
>
>
> Best wishes
>
> Jakob
>
>
>
> Jakob Ribbing, Ph.D.
>
> Senior Consultant, Pharmetheus AB
>
>
> Cell/Mobile: +46 (0)70 514 33 77
>
> [email protected]
>
> www.pharmetheus.com
>
> Phone, Office: +46 (0)18 513 328
>
> Uppsala Science Park, Dag Hammarskjölds väg 36B
> SE-752 37 Uppsala, Sweden
>
>
>
>
> On 3 May 2021, at 18:45, Shankar Lanke <[email protected]> wrote:
>
> Dear NMUsers,
> I am having trouble with PSN4.9, if anyone has a similar issue may I know
> how to fix it?
> I ran a model that was developed prior to the release of PSN4.9. In PSN4.9
> the covariance step is aborted because S Matrix is unobtainable. I purged
> 4.9 and loaded 4.8, the model ran perfectly and matched the
> previous results. Thank you very much in advance.
>
>
> *Version 4.9*
> #TERE:
> Elapsed estimation time in seconds: 633.22
> 0S MATRIX UNOBTAINABLE
> 0COVARIANCE STEP ABORTED
> Elapsed covariance time in seconds: 47.98
> Elapsed postprocess time in seconds: 51.79
>
>
> *Version 4.8*
> #TERM:
> 0MINIMIZATION SUCCESSFUL
> NO. OF FUNCTION EVALUATIONS USED: 783
> NO. OF SIG. DIGITS IN FINAL EST.: 2.0
>
>
> Regards,
> Shankar Lanke
>
>
>
> *This communication is confidential and is only intended for the use of
> the individual or entity to which it is directed. It may contain
> information that is privileged and exempt from disclosure under applicable
> law. If you are not the intended recipient please notify us immediately.
> Please do not copy it or disclose its contents to any other person.*
>
--
Regards,
Shankar Lanke Ph.D.