[gpaw-users] Error in gpaw transport

Georg Kastlunger georg.kast at hotmail.com
Thu Oct 3 12:09:32 CEST 2013


Hi Jingzhe,

I have now tried to do the calculation using pl_kpts=(1,1,13), which 
unfortunately this didnt work either.

I habe noticed that you are defining the pl kpoints in a list (i.e. 
pl_kpts=[1,1,13] ), while I am defining them in a tuple as shown above.
Could this be the reason of my problem?

Best wishes,
Georg


On 10/02/2013 02:03 PM, jingzhe Chen wrote:
> Hi Georg,
>
>          If you do it in parallel, the K points in Z direction are 
> distributed with such
> setting. As said in the description of pl_kpts in the wiki, we want to 
> keep the numbers
> of k points in X and Y direction same in scattering region and leads, 
>  and the only K
> points in these two directions are distributed, and the safe way is to 
> use a prime
> number in the Z direction of pl_kpts. In your case, choosing 
> pl_kpts=[1,1,13] or [1,1,17]
> should be ok.
>
>         Best.
>         Jingzhe
>  ,
>
>
>
> On Wed, Oct 2, 2013 at 6:36 AM, Georg Kastlunger 
> <georg.kast at hotmail.com <mailto:georg.kast at hotmail.com>> wrote:
>
>     Dear mailing list,
>
>     recently I have tried to apply the transport routine implemented
>     in gpaw. Unfortunately I experience an error which I don't really
>     understand.
>     The error is the following:
>
>     Traceback (most recent call last):
>       File "gpaw_transport.py", line 36, in <module>
>         t.calculate_iv()
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/calculator.py",
>     line 2402, in calculate_iv
>         self.negf_prepare()
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/calculator.py",
>     line 1239, in negf_prepare
>         self.initialize_transport()
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/calculator.py",
>     line 402, in initialize_transport
>         self.calculate_leads()
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/calculator.py",
>     line 342, in calculate_leads
>         self.collect_leads_matrices(atoms.calc, i)
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/calculator.py",
>     line 1098, in collect_leads_matrices
>         direction=lead_direction)
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/tools.py",
>     line 344, in get_pk_hsd
>         hl_spkmm = substract_pk(d, npk, ntk, kpts, hl_skmm, hors='h',
>     magnet=magnet)
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/tools.py",
>     line 215, in substract_pk
>         tkpts = pick_out_tkpts(d, npk, ntk, kpts)
>       File
>     "/home/lv70174/georg/gpaw_inst_10428/lib64/python2.6/site-packages/gpaw/transport/tools.py",
>     line 232, in pick_out_tkpts
>         tkpts[i, d] = kpts[i, d]
>     IndexError: index (8) out of range (0<=index<8) in dimension 0
>
>     As far as I understand from the code the index d defines the axis
>     and i is the number of kpoints in this specific direction. Am I
>     wrong with that?
>
>     The calculator I am defining uses 1 kpoint in the scattering
>     region and 1,1,15 in the leads as identical as in the tutorial:
>
>     t = Transport(h=0.18,
>                   xc='PBE',
>                   kpts=(1,1,1),
>                   occupations=FermiDirac(0.1),
>                   mode='lcao',
>                   basis='dzp',
>                   poissonsolver=PoissonSolver(nn=2, relax='GS'),
>                   txt='transport.txt',
>                   mixer=Mixer(0.05, 5, weight=100.0),
>                   pl_atoms=[pl_atoms1, pl_atoms2],
>                   pl_cells=[pl_cell1, pl_cell2],
>                   pl_kpts=(1,1,16),
>                   non_sc=True)
>
>     Has anyone experienced this problem and knows how to solve it? I
>     would be very grateful if someone could help me with that.
>
>     Best regards,
>     Georg
>
>     _______________________________________________
>     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
>
>
>
>
> _______________________________________________
> gpaw-users mailing list
> 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/20131003/20da8b9f/attachment.html>


More information about the gpaw-users mailing list