search

UUPharmacometrics / PsN

Perl-Speaks-NONMEM
https://uupharmacometrics.github.io/PsN
GNU General Public License v2.0
64 stars 19 forks source link

NMtran could not be initiated (the NMtran output file FDATA is missing) #257

Open liangqiongyue opened 4 months ago

liangqiongyue commented 4 months ago 
          Make sure that you can run NONMEM correctly without PsN, i.e. try ```C:\nm7C\ver7_20100223\gf\ref\run\nmfe7 run1.mod run1.lst```

Originally posted by @rikardn in https://github.com/UUPharmacometrics/PsN/issues/141#issuecomment-522545633

liangqiongyue commented 4 months ago

Hello, I encountered the same problem. I followed your prompts and ran the nmfe7 run1.mod run1.lst command. You can see that it can run smoothly.

image

Please tell me what should I do next.

Thanks

liangqiongyue commented 4 months ago

image

rikardn commented 4 months ago

Something is fundamentally wrong with running your model. Check the run directory and see what files were generated in the NM_run directories.

Another thing: I would not use the parafile in this case. It will mix two methods of parallelization. Either use only threads or parafile together with nodes. This could even be the source of the error.

liangqiongyue commented 3 days ago

Hi rikardn, I utilize the '-parafile=pirana_auto_mpi. pnm' option because I intend to leverage multi-core processing to expedite the model's execution time.

Following your guidance, I have combined the '-parafile=pirana_auto_mpi. pnm' with the '-nodes=32' directive, but I am encountering the same error. What steps can I take to resolve this error while still employing multi-core execution?"

微信图片_20240703153149

Something is fundamentally wrong with running your model. Check the run directory and see what files were generated in the NM_run directories.

Another thing: I would not use the parafile in this case. It will mix two methods of parallelization. Either use only threads or parafile together with nodes. This could even be the source of the error.