Home | Installation | Control Streams | Bootstrap | Randomization Test | Visual Predictive Check | Autocovariate | Files | References
Last Updated: 3 April 2012
WFN provides a simple way to switch between different installation versions of NONMEM. An installation version is not a full version like NONMEM V or NONMEM VI but is a specific installation of one of the full versions. Different installation versions are created to increase the size of NONMEM limits e.g. number of subjects, number of parameters, number of observations per subject.
The simplest way to create WFN installation versions is to use the NMQWFN package which uses the NMQUAL software developed by Metrum Institute. Four installation versions are created (std, s70, 500, 570).
A WFN utility package has been created to allow testing and comparison of performance when using the different installation versions. The package uses vercom.bat to loop through the different installation versions and through one or more compilers that you may have used to create these versions. Addtime.awk is used by vercom.bat to add the runtime of each run to the results file.
Three example control streams/data sets are supplied. These are theopd_std, ka1_std and ka1_emax1_std. Theopd_std is a simple pharmacodynamic model, ka1-std is a first order absorption and first order elimination model for warfarin and ka1_emax1_std is a PKPD model using the same warfarin PK as used in ka1_std.
You can run these problems as follows:
vercom theopd_std
vercom ka1_std
vercom ka1_emax1_std
The results are created in files called theopd_std_vercom.txt, ka1_std_vercom.txt, ka1_emax1_std_vercom.txt. These are tab delimited files that can easily be loaded into Excel. Examples are shown with theopd_std and ka1_emax1_std obtained using Intel Fortran 9.1. Notice how the run times (min:sec) increase as the installation version is made more complex. This is less prominent for the run with longer run times (over 2 mins instead of under 10 seconds).
|
#Run |
RunTime |
Obj |
Min |
Cov |
Eval |
Sig |
Sub |
Obs |
POPE0 |
POPEMAX |
POPEC50 |
ETAE0 |
ETAEMAX |
ETAEC50 |
ERRSD |
POPE0:se |
POPEMAX:se |
POPEC50:se |
ETAE0:se |
ETAEMAX:se |
ETAEC50:se |
ERRSD:se |
|
theopd_std_ivf_570 |
00:09.9 |
5801.321 |
MINIMIZATION_SUCCESSFUL |
OK |
275 |
4.2 |
153 |
574 |
139 |
191 |
8.8 |
0.098641 |
0.267769 |
1.46287 |
81.0555 |
7.9 |
12.9 |
17.8 |
355.6 |
42.4 |
34.3 |
10.7 |
|
theopd_std_ivf_500 |
00:08.5 |
5801.321 |
MINIMIZATION_SUCCESSFUL |
OK |
275 |
4.2 |
153 |
574 |
139 |
191 |
8.8 |
0.098641 |
0.267769 |
1.46287 |
81.0555 |
7.9 |
12.9 |
17.8 |
355.6 |
42.4 |
34.3 |
10.7 |
|
theopd_std_ivf_s70 |
00:05.4 |
5801.321 |
MINIMIZATION_SUCCESSFUL_ |
OK |
275 |
4.2 |
153 |
574 |
139 |
191 |
8.8 |
0.098641 |
0.267769 |
1.46287 |
81.0555 |
7.9 |
12.9 |
17.8 |
355.6 |
42.4 |
34.3 |
10.7 |
|
theopd_std_ivf_std |
00:04.1 |
5801.321 |
MINIMIZATION_SUCCESSFUL |
OK |
275 |
4.2 |
153 |
574 |
139 |
191 |
8.8 |
0.098641 |
0.267769 |
1.46287 |
81.0555 |
7.9 |
12.9 |
17.8 |
355.6 |
42.4 |
34.3 |
10.7 |
|
#Run |
RunTime |
Obj |
Min |
Cov |
Eval |
Sig |
Sub |
Obs |
POPCL |
POPV |
POPTABS |
POPLAG |
POPE0 |
POPEMAX |
POPC50 |
POPTEQ |
PPVCL |
PPVV |
PPVTABS |
PPVLAG |
PPVE0 |
PPVEMAX |
PPVC50 |
PPVTEQ |
RUVCV |
RUVSD |
RUVFX |
POPCL:se |
POPV:se |
POPTABS:se |
POPLAG:se |
POPE0:se |
POPEMAX:se |
POPC50:se |
POPTEQ:se |
PPVCL:se |
PPVV:se |
PPVTABS:se |
PPVLAG:se |
PPVE0:se |
PPVEMAX:se |
PPVC50:se |
PPVTEQ:se |
RUVCV:se |
RUVSD:se |
RUVFX:se |
|
ka1_emax1_std_ivf_570 |
02:30.5 |
1205.269 |
MINIMIZATION_SUCCESSFUL |
OK |
145 |
3.3 |
32 |
482 |
0.134 |
8.1 |
0.523 |
0.822 |
96.6 |
-1 |
1.18 |
13 |
0.267021 |
0.134536 |
0.834266 |
0.582237 |
0.053759 |
0 |
0.446094 |
0.10583 |
0.086718 |
0.257099 |
3.84708 |
0 |
0 |
0 |
0 |
1.2 |
0 |
8.9 |
2.7 |
0 |
0 |
0 |
0 |
34.9 |
0 |
22.5 |
44.3 |
0 |
0 |
18.4 |
|
ka1_emax1_std_ivf_500 |
02:26.0 |
1205.269 |
MINIMIZATION_SUCCESSFUL |
OK |
145 |
3.3 |
32 |
482 |
0.134 |
8.1 |
0.523 |
0.822 |
96.6 |
-1 |
1.18 |
13 |
0.267021 |
0.134536 |
0.834266 |
0.582237 |
0.053759 |
0 |
0.446094 |
0.10583 |
0.086718 |
0.257099 |
3.84708 |
0 |
0 |
0 |
0 |
1.2 |
0 |
8.9 |
2.7 |
0 |
0 |
0 |
0 |
34.9 |
0 |
22.5 |
44.3 |
0 |
0 |
18.4 |
|
ka1_emax1_std_ivf_s70 |
02:22.2 |
1205.269 |
MINIMIZATION_SUCCESSFUL |
OK |
145 |
3.3 |
32 |
482 |
0.134 |
8.1 |
0.523 |
0.822 |
96.6 |
-1 |
1.18 |
13 |
0.267021 |
0.134536 |
0.834266 |
0.582237 |
0.053759 |
0 |
0.446094 |
0.10583 |
0.086718 |
0.257099 |
3.84708 |
0 |
0 |
0 |
0 |
1.2 |
0 |
8.9 |
2.7 |
0 |
0 |
0 |
0 |
34.9 |
0 |
22.5 |
44.3 |
0 |
0 |
18.4 |
|
ka1_emax1_std_ivf_std |
02:23.2 |
1205.269 |
MINIMIZATION_SUCCESSFUL |
OK |
145 |
3.3 |
32 |
482 |
0.134 |
8.1 |
0.523 |
0.822 |
96.6 |
-1 |
1.18 |
13 |
0.267021 |
0.134536 |
0.834266 |
0.582237 |
0.053759 |
0 |
0.446094 |
0.10583 |
0.086718 |
0.257099 |
3.84708 |
0 |
0 |
0 |
0 |
1.2 |
0 |
8.9 |
2.7 |
0 |
0 |
0 |
0 |
34.9 |
0 |
22.5 |
44.3 |
0 |
0 |
18.4 |
Home | Installation | Control Streams | Bootstrap | Randomization Test | Visual Predictive Check | Autocovariate | Files | References