Author Topic: Error in mobility calculation  (Read 1264 times)

0 Members and 1 Guest are viewing this topic.

Offline arlonne

  • New ATK user
  • *
  • Posts: 9
  • Country: cn
  • Reputation: 0
    • View Profile
Error in mobility calculation
« on: April 6, 2017, 15:26 »
Hi, everyone

I need your great help. Thank you.

version: 2016.4

last lines in output file:
"ElectronPhononCoupling: Done with 84 % of the calculations. ETC 0:00:00
ElectronPhononCoupling: Done with 92 % of the calculations. ETC 0:00:00
Mobility: Cannot refine data. Continuing without refinement.
Mobility: Cannot refine data. Continuing without refinement.
Mobility: Calculating transition rates for phonon mode: 1  (1 / 17)
Mobility: Cannot refine data. Continuing without refinement.
Mobility: Calculating transition rates for phonon mode: 0  (0 / 17)
Mobility: Calculating transition rates for phonon mode: 2  (2 / 17)
Mobility: Cannot refine data. Continuing without refinement."

error info:
"Traceback (most recent call last):
  File "AlPS4-vpopt-mobility.py", line 408, in <module>
    refinement=1,
  File "./zipdir/NL/Analysis/Mobility.py", line 194, in __init__
Traceback (most recent call last):
  File "AlPS4-vpopt-mobility.py", line 408, in <module>
    refinement=1,
  File "./zipdir/NL/Analysis/Mobility.py", line 194, in __init__
  File "./zipdir/NL/Analysis/Mobility.py", line 970, in _calculateTransitionRate
  File "./zipdir/NL/ComputerScienceUtilities/ParallelTools/ParallelTools.py", line 390, in parallelTask
  File "./zipdir/NL/ComputerScienceUtilities/ParallelTools/StaticTaskScheduler.py", line 212, in run
  File "./zipdir/NL/ComputerScienceUtilities/ParallelTools/StaticTaskScheduler.py", line 269, in __runInParallelMode
  File "./zipdir/NL/Analysis/Mobility.py", line 951, in calculatePSpinModeKQ
  File "./zipdir/NL/Analysis/Mobility.py", line 613, in _energiesKQ
  File "./zipdir/NL/Analysis/Mobility.py", line 2157, in _interpolateEnergies
AttributeError: 'numpy.ndarray' object has no attribute 'inUnitsOf'
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 1"

Offline Ulrik G. Vej-Hansen

  • QuantumATK Staff
  • Supreme ATK Wizard
  • *****
  • Posts: 411
  • Country: dk
  • Reputation: 7
    • View Profile
Re: Error in mobility calculation
« Reply #1 on: April 7, 2017, 09:11 »
Please provide the script as well.

Offline arlonne

  • New ATK user
  • *
  • Posts: 9
  • Country: cn
  • Reputation: 0
    • View Profile
Re: Error in mobility calculation
« Reply #2 on: April 7, 2017, 10:17 »
Thank you. pls see the attachment.

Offline Ulrik G. Vej-Hansen

  • QuantumATK Staff
  • Supreme ATK Wizard
  • *****
  • Posts: 411
  • Country: dk
  • Reputation: 7
    • View Profile
Re: Error in mobility calculation
« Reply #3 on: April 7, 2017, 11:58 »
There seems to be an error in your q-point selection: (0,0,10) means that you sample q=0 10 times, which I guess is not what you want. This could be the cause of an error further on. Please see the tutorial for more information on how to set the q-points: http://docs.quantumwise.com/tutorials/mobility/mobility.html

I will try to reproduce.

But it also looks like this script is not the one that actually produced the error. In the traceback, it looks like the error happens in line 408, but the attached script only contains 403 lines.

Offline arlonne

  • New ATK user
  • *
  • Posts: 9
  • Country: cn
  • Reputation: 0
    • View Profile
Re: Error in mobility calculation
« Reply #4 on: April 10, 2017, 08:39 »

The difference is because I have run this for two times.  The script I gave to you is used in the first running, that is from the scratch. Because it has errors as I posted, then I read the ground state from the first one and performed the second running. Although the error info I posted is from the second runing, it is same to the first running.

And I will check the q-point carefully. Thanks again.