[gpaw-users] calc.get_pseudo_wave_function error in tutorial "Creating wave function cube files"

Jens Jørgen Mortensen jensj at fysik.dtu.dk
Fri Aug 1 15:26:11 CEST 2014


On 08/01/2014 03:10 PM, Pu ZHANG wrote:
> I'm sorry I'm not sure which script you are referring to. Do you mean 
> the script used for compilation?

No, the cubeWF.py script will read a gpw-file and write cube files (I'm 
guessing).  To debug this, I would need your gpw-file or a script that 
can calculate it.

Jens Jørgen

> Then I'm using the gpaw compiled previously by our cluster 
> administrator. I don't know the details.
>
> Thanks!
>
>
>     Best regards, Pu Zhang
>
> --
> Postdoc
> The group of Structured Electromagnetic Materials
> Department of Photonics Engineering
> Technical University of Denmark
> Ørsteds Plads, Building 345v, 166
> 2800, Kgs. Lyngby
> E-mail: puzha at fotonik.dtu.dk <mailto:puzha at fotonik.dtu.dk>
> Homepage: http://www.dtu.dk/Service/Telefonbog/Person?id=91944&tab=0
> Phone: 45 25 36 43
> Fax: 45 93 65 81
>
>
> On Fri, Aug 1, 2014 at 3:03 PM, Jens Jørgen Mortensen 
> <jensj at fysik.dtu.dk <mailto:jensj at fysik.dtu.dk>> wrote:
>
>
>     On 08/01/2014 02:56 PM, Pu ZHANG wrote:
>>     Hi, Jens!
>>
>>     Thanks a lot for the reply!
>>
>>     Yes, I was running in parallel. And it does work if I use only
>>     one processor. The script is as follows:
>
>     To fix this when it runs in parallel, we would need to see the
>     script that generated the gpw-file.
>
>     Jens Jørgen
>
>
>>
>>     job script
>>     -------------
>>     #!/bin/bash
>>     #PBS -N cubeWF
>>     #PBS -e $PBS_JOBNAME.$PBS_JOBID.err
>>     #PBS -o $PBS_JOBNAME.$PBS_JOBID.log
>>     #PBS -q fotonano
>>     #PBS -A fotonano
>>     #PBS -l nodes=1:ppn=2
>>     #PBS -l walltime=02:00:00
>>
>>     source $PBS_O_HOME/.bashrc
>>     module purge
>>     module load npa-cluster-setup
>>
>>     ulimit -s unlimited
>>     cd $PBS_O_WORKDIR
>>
>>     module load python2.7.numpy.scipy.scientific.ase.gpaw.inelastica
>>     module load gpaw-setups/0.9.9672
>>     xc=`which xc-info`
>>     xc_h=$(dirname $(dirname $xc))
>>     export LD_LIBRARY_PATH=$xc_h/lib:$LD_LIBRARY_PATH
>>
>>     m=`which mpirun`
>>     m_h=$(dirname $(dirname $m))
>>     export LD_LIBRARY_PATH=$m_h/lib:$LD_LIBRARY_PATH
>>
>>     export GPAW_FFTWSO="libmkl_intel_lp64.so"
>>
>>     mpirun -np $PBS_NP gpaw-python "cubeWF.py" > RUN.out
>>     -------------
>>
>>
>>         Best regards, Pu Zhang
>>
>>     --
>>     Postdoc
>>     The group of Structured Electromagnetic Materials
>>     Department of Photonics Engineering
>>     Technical University of Denmark
>>     Ørsteds Plads, Building 345v, 166
>>     2800, Kgs. Lyngby
>>     E-mail: puzha at fotonik.dtu.dk <mailto:puzha at fotonik.dtu.dk>
>>     Homepage: http://www.dtu.dk/Service/Telefonbog/Person?id=91944&tab=0
>>     Phone: 45 25 36 43
>>     Fax: 45 93 65 81
>>
>>
>>     On Fri, Aug 1, 2014 at 2:38 PM, Jens Jørgen Mortensen
>>     <jensj at fysik.dtu.dk <mailto:jensj at fysik.dtu.dk>> wrote:
>>
>>         On 07/28/2014 05:13 PM, Pu ZHANG wrote:
>>>         From the error information it seems something's wrong with
>>>         the argument when calling the
>>>         function get_pseudo_wave_function. But I searched around and
>>>         couldn't figure out the problem. Have anyone ever had such a
>>>         problem?
>>
>>         Are you running this in parallel?  If yes - then try running
>>         it in serial. Could you show us your input script?
>>
>>         Jens Jørgen
>>
>>>
>>>         Thanks!
>>>
>>>
>>>             Best regards, Pu Zhang
>>>
>>>         --
>>>         Postdoc
>>>         The group of Structured Electromagnetic Materials
>>>         Department of Photonics Engineering
>>>         Technical University of Denmark
>>>         Ørsteds Plads, Building 345v, 166
>>>         2800, Kgs. Lyngby
>>>         E-mail: puzha at fotonik.dtu.dk <mailto:puzha at fotonik.dtu.dk>
>>>         Homepage:
>>>         http://www.dtu.dk/Service/Telefonbog/Person?id=91944&tab=0
>>>         Phone: 45 25 36 43
>>>         Fax: 45 93 65 81
>>>
>>>
>>>         On Fri, Jul 25, 2014 at 3:06 PM, Pu ZHANG
>>>         <puzhang0702 at gmail.com <mailto:puzhang0702 at gmail.com>> wrote:
>>>
>>>             Dear all,
>>>
>>>             Execution of the tutorial example "Creating wave
>>>             function cube files" leads to error related
>>>             to calc.get_pseudo_wave_function.
>>>
>>>             The script is
>>>             https://wiki.fysik.dtu.dk/gpaw/tutorials/plotting/plot_wave_functions.html,
>>>             and the error message looks like:
>>>
>>>             Traceback (most recent call last):
>>>               File "cubeWF.py", line 14, in <module>
>>>                 wf = calc.get_pseudo_wave_function(band=band)
>>>               File
>>>             "/zdata/groups/common/nicpa/2014-may/XeonX5550/python/2.7.6/packages/gpaw/0.10.0.11364/intel-14.0.1/lib/python2.7/site-packages/gpaw/aseinterface.py",
>>>             line 407, in get_pseudo_wave_function
>>>                 pad=False)
>>>               File
>>>             "/zdata/groups/common/nicpa/2014-may/XeonX5550/python/2.7.6/packages/gpaw/0.10.0.11364/intel-14.0.1/lib/python2.7/site-packages/gpaw/aseinterface.py",
>>>             line 412, in get_pseudo_wave_function
>>>                 psit_G = self.wfs.get_wave_function_array(band, kpt,
>>>             spin)
>>>               File
>>>             "/zdata/groups/common/nicpa/2014-may/XeonX5550/python/2.7.6/packages/gpaw/0.10.0.11364/intel-14.0.1/lib/python2.7/site-packages/gpaw/wavefunctions/base.py",
>>>             line 401, in get_wave_function_array
>>>                 psit_G = self.gd.collect(psit_G)
>>>               File
>>>             "/zdata/groups/common/nicpa/2014-may/XeonX5550/python/2.7.6/packages/gpaw/0.10.0.11364/intel-14.0.1/lib/python2.7/site-packages/gpaw/grid_descriptor.py",
>>>             line 459, in collect
>>>                 A_xg[..., b0:e0, b1:e1, b2:e2] = a_xg
>>>             ValueError: could not broadcast input array from shape
>>>             (23,23,23) into shape (11,23,23)
>>>             GPAW CLEANUP (node 0): <type 'exceptions.ValueError'>
>>>             occurred.  Calling MPI_Abort!
>>>             [n-62-13-7:00605] 1 more process has sent help message
>>>             help-mpi-runtime.txt / mpi_init:warn-fork
>>>             [n-62-13-7:00605] Set MCA parameter
>>>             "orte_base_help_aggregate" to 0 to see all help / error
>>>             messages
>>>
>>>             Any idea? Is there extra information shall I provide to
>>>             help diagnose the problem?
>>>
>>>             Thanks!
>>>
>>>
>>>                 Best regards, Pu Zhang
>>>
>>>             --
>>>             Postdoc
>>>             The group of Structured Electromagnetic Materials
>>>             Department of Photonics Engineering
>>>             Technical University of Denmark
>>>             Ørsteds Plads, Building 345v, 166
>>>             2800, Kgs. Lyngby
>>>             E-mail: puzha at fotonik.dtu.dk <mailto:puzha at fotonik.dtu.dk>
>>>             Homepage:
>>>             http://www.dtu.dk/Service/Telefonbog/Person?id=91944&tab=0
>>>             Phone: 45 25 36 43 <tel:45%2025%2036%2043>
>>>             Fax: 45 93 65 81 <tel:45%2093%2065%2081>
>>>
>>>
>>>
>>>
>>>         _______________________________________________
>>>         gpaw-users mailing list
>>>         gpaw-users at listserv.fysik.dtu.dk  <mailto:gpaw-users at listserv.fysik.dtu.dk>
>>>         https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>>
>>
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20140801/b2af674a/attachment-0001.html>


More information about the gpaw-users mailing list