Hi Pauline,
There seems to a be a loophole in the code. A user should never get an
assertion, there should be a runtime error. I notice that you are using
arts-2.0, so maybe this loophole is fixed in the development version
(2.1) ...
For everybody:
Some general hints about how to provide some more information if asserts
like this is occuring. At least if you are running lunix/unix, and has
the debugger installed.
Defualt in atmlab is to remove the "work directory" (the one in /tmp).
To keep these directories, do in matlab
>> atmlab('DEBUG',true)
Run your case, and efter the assert in a terminal go to the work folder
listed in the error message, start the debugger and run arts. An example:
cd /tmp/atmlab-martinet-tp897f84cb_6b6b_4953_bc99_95e5f8f4ad7
gdb arts
run cfile_y.arts
where
The last two lines are called in the debugger. Check the error message
to call the correct cfile (cfile_XX.arts). You can also need to call
down or up after "where" to locate the interesting part in call stack.
The interesting part has normally levels 5 - 10. Cpy this part and
include in email.
After this, remove all temporary folders created in /tmp, and reset
>> atmlab('DEBUG',false)
Pauline:
Maybe you can try the above. Alternatively, can you set up a small test
case that generates the error? That I easily can set-up here on my side.
Bye,
Patrick
On 03/28/14 17:15, MARTINET Pauline wrote:
> Hi,
>
> I have juste started to work with the ARTS model and the qpack2
> package to perform temperature and humidity retrievals.
>
> I think I managed to run qpack2 for the temperature (I still have some
> problems with humidity).
>
> I have a loop over about 200 atmospheric profiles and I retrieve the
> temperature profile from these
> atmospheric profiles and the observation from a ground-based microwave
> radiometer.
>
> However for some profiles I have the following error that I do not
> understand:
>
> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
> arts: /home/martinet/RADIOMETRE/arts-2.0.41/src/interpolation.cc:202:
> void gridpos(ArrayOfGridPos&, ConstVectorView, ConstVectorView, const
> Numeric&): Assertion `is_increasing(old_grid)' failed.
> /home/martinet/RADIOMETRE/arts-2.0.41/build/src/arts -r00
> /tmp/atmlab-martinet-tp897f84cb_6b6b_4953_bc99_95e5f8f4ad7f/cfile_y.arts:
> Aborted
>
> Error using arts (line 47)
> An error occured while executing ARTS. See above.
>
> Error in arts_oem (line 43)
> arts( cfile );
>
> Error in oem (line 785)
> [R,yf] = feval( comfun, Q, R, x, iter );
>
> Error in qpack2 (line 114)
> [X,R] = oem( O, Qm, R, @arts_oem, Sx, Se, Sxinv, [], xa, Y(m).Y );
>
> Error in retrieval_profiles (line 132)
> L2 = qpack2( Q, O, Y );
>
>
> -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>
> This message appears after at least one iteration.
>
> Does anyone know what is wrong?
>
> Thank you very much for your help.
>
> Best regards,
>
> Pauline
>
>
> ----- Météo-France -----
> MARTINET PAULINE
> CNRM/GMEI/LISA
> [email protected]
> Fixe : +33 561079031
>
>
>
> _______________________________________________
> qpack mailing list
> [email protected]
> https://www.sat.ltu.se/mailman/listinfo/qpack
>
_______________________________________________
qpack mailing list
[email protected]
https://www.sat.ltu.se/mailman/listinfo/qpack
