CalledProcessError

56 views
Skip to first unread message

avvalm...@gmail.com

unread,
May 1, 2019, 8:47:00 AM5/1/19
to message_ix
Hello, I am trying to develop India energy system model from the Austria tutorial, however I get an error when runnin the cell 'scenario.solve()':

CalledProcessError: Command 'gams "C:\path\to\model\MESSAGE_run.gms" --in="C:\path\to\model\data\MsgData_India_energy_model_baseline.gdx" --out="C:\path\to\model\output\MsgOutput_India_energy_model_baseline.gdx" --iter="C:\path\to\model\output\MsgIterationReport_India_energy_model_baseline.gdx" LogOption=4' returned non-zero exit status 3.

In Anaconda Prompt I get this:

Iteration log . . .
Iteration:     1    Scaled infeas =          1873.950799
Iteration:    78    Scaled infeas =            64.616511
LP status(3): infeasible
Cplex Time: 0.02sec (det. 0.62 ticks)

Model has been proven infeasible.
--- Restarting execution
--- MESSAGE_run.gms(3803) 2 Mb
--- Reading solution for model MESSAGE_LP
--- MESSAGE_run.gms(3785) 3 Mb
    +++ MESSAGEix did not solve to optimality - run is aborted, no output produced! +++
--- MESSAGE_run.gms(3785) 3 Mb 1 Error
*** Status: Execution error(s)
--- Job MESSAGE_run.gms Stop 05/01/19 14:31:42 elapsed 0:00:00.464
[I 14:33:06.228 NotebookApp] Saving file at /message/tutorial/Austrian_energy_system/austria.ipynb

Does someone know why I get it? Thanks in advance.

Matthew Gidden

unread,
May 2, 2019, 3:32:40 AM5/2/19
to avvalm...@gmail.com, message_ix
Hi!

The critical line is: 
Model has been proven infeasible.
Thus, some part of the parameterization (i.e., a constraint coefficient or something else) causes the model to be infeasible. There should be a ".lst" file in the same directory as the MESSAGE GAMS code which lists infeasibilities with four stars (****).  This website might be able to help a bit: http://yetanothermathprogrammingconsultant.blogspot.com/2018/08/the-best-way-to-debug-infeasible-models.html

If you would like, you could make an issue on github.com/iiasa/message_ix to print these lines in the error message when a model is infeasible. We can then put this feature on our radar for an upcoming release.

Best,
Matt 



--
You received this message because you are subscribed to the Google Groups "message_ix" group.
To unsubscribe from this group and stop receiving emails from it, send an email to message_ix+...@googlegroups.com.
To post to this group, send email to messa...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/message_ix/6593a38a-d0b4-44a8-b317-024aeacb751d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
Reply all
Reply to author
Forward
0 new messages