QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: NaKa on January 25, 2018, 15:23

Title: An error about 'mulliken-analysis.py' document
Post by: NaKa on January 25, 2018, 15:23

Hi,

I tried to analyse .nc file of my run according to 'mulliken-analysis.py' document given on page Atomic Scale Capacitance, but have some problems.

Actually, I did it before by an older version of VNL, I guess 2015.  Now I try by 2017 version and get an error message like;

Traceback (most recent call last):
  File "ANALYSIS_mulliken.py", line 26, in <module>
    initial_population = realVectorToNumpy(builder.neutralAtomicOccupations())
  File "zipdir/NL/ComputerScienceUtilities/Functions.py", line 34, in realVectorToNumpy
TypeError: in method 'realVectorToNumpy', argument 1 of type 'RealVector const &'[/size]

Relevant part of the script is;
"# Get the atomic occupations from the zero-bias config
calculator = configuration_list[0].calculator()
builder = calculator._builder()(configuration_list[0], calculator)
initial_population = realVectorToNumpy(builder.neutralAtomicOccupations())"

Can you please help, how can I fixed it?

Title: Re: An error about 'mulliken-analysis.py' document
Post by: Petr Khomyakov on January 26, 2018, 10:00

Do I understand correctly that you have been trying reproducing the results in this tutorial https://docs.quantumwise.com/tutorials/atomic-scale_capacitance/atomic-scale_capacitance.html, and you have used the  'mulliken-analysis.py' script from that tutorial in ATK-2017?

Title: Re: An error about 'mulliken-analysis.py' document
Post by: NaKa on January 26, 2018, 12:16

Yes it is right. Actually I have tried it for a similar structure, set by gold atoms and graphene. Does it matter?