QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: njuxyh on April 29, 2014, 21:38

Title: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: njuxyh on April 29, 2014, 21:38

Hi, after i optimize the central region of a device,  i only found the nc file contains only GI000 not GID001,
the optimization part of my script is :
...........
...........

 #----------------------------------------
# Device Calculator
#----------------------------------------
calculator = DeviceLCAOCalculator(
    basis_set=basis_set,
    exchange_correlation=exchange_correlation,
    numerical_accuracy_parameters=device_numerical_accuracy_parameters,
    iteration_control_parameters=device_iteration_control_parameters,
    poisson_solver=device_poisson_solver,
    dynamical_matrix_parameters=device_dynamical_matrix_parameters,
    contour_parameters=contour_parameters,
    device_algorithm_parameters=device_algorithm_parameters,
    electrode_calculators=
        [left_electrode_calculator, right_electrode_calculator],
    electrode_voltages=( 0.0*Volt, 0.0*Volt)
    )

device_configuration.setCalculator(calculator)
nlprint(device_configuration)
device_configuration.update()
nlsave('au-a-5-h-au.nc', device_configuration, labels=['relax'])

constraints = [FixStrain(x=True, y=True, z=False)]
device_configuration = OptimizeGeometry(
        device_configuration,
        max_forces=0.05*eV/Ang,
        max_stress=0.05*eV/Ang**3,
        max_steps=200,
        max_step_length=0.5*Ang,
        constraints=constraints,
        trajectory_filename=None,
        optimizer_method=QuasiNewton(),
        )
nlsave('au-molecular-au.nc', device_configuration, labels=['relax'])
nlprint(device_configuration)

so my puzzle is why the atoms are not relaxed?
and after the job, how i see the finally geometry of central region, including atomic position.

i need your help!

thanks

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: Anders Blom on April 29, 2014, 22:52

You are saving the optimized geometry to a different file

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: njuxyh on April 29, 2014, 23:24

NO, i am very sorry!

in the first post,i have written wrong, the practical script is here, and the vnl show is in attachment, the nc file indeed only have GID000, not GID001


 #----------------------------------------
# Device Calculator
#----------------------------------------
calculator = DeviceLCAOCalculator(
    basis_set=basis_set,
    exchange_correlation=exchange_correlation,
    numerical_accuracy_parameters=device_numerical_accuracy_parameters,
    iteration_control_parameters=device_iteration_control_parameters,
    poisson_solver=device_poisson_solver,
    dynamical_matrix_parameters=device_dynamical_matrix_parameters,
    contour_parameters=contour_parameters,
    device_algorithm_parameters=device_algorithm_parameters,
    electrode_calculators=
        [left_electrode_calculator, right_electrode_calculator],
    electrode_voltages=( 0.0*Volt, 0.0*Volt)
    )

device_configuration.setCalculator(calculator)
nlprint(device_configuration)
device_configuration.update()
nlsave('au-a-5-h-au-optimization.nc', device_configuration)

constraints = [FixStrain(x=True, y=True, z=False)]
device_configuration = OptimizeGeometry(
        device_configuration,
        max_forces=0.1*eV/Ang,
        max_stress=0.01*eV/Ang**3,
        max_steps=200,
        max_step_length=0.5*Ang,
        constraints=constraints,
        trajectory_filename='relax-process',
        optimizer_method=QuasiNewton(),
        )
nlsave('au-a-5-h-au-optimization.nc', device_configuration, labels=['relaxed-geometry'])
nlprint(device_configuration)

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: Anders Blom on April 30, 2014, 00:03

I think this is because your trajectory filename doesn't have the .nc extension, so the script actually terminated and never ran the optimization. There should be some error message about that in the log file?

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: njuxyh on April 30, 2014, 00:20

yes, i saw in the end of log file :

=====================================================================================
=   BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
=   EXIT CODE: 256
=   CLEANING UP REMAINING PROCESSES
=   YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
=====================================================================================


is that cause the optimization do not run at all?

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: Anders Blom on April 30, 2014, 12:33

That message is not from ATK, it's from the MPI system. But it can be caused by the fact that there is an error in ATK, such as for instance - as I suspect - the lacking extension on the trajectory file.

Title: Re: GID000 Came out but GID001 does not appear (VERSION 13.8.1)
Post by: njuxyh on April 30, 2014, 17:39

Hi Prof. Anders Blom:

after your reply, i modify the script, add the trajectory_filename='relax.nc', but the problem is still and complicated than before.

Because the outfile is so large, i email them to you, please check your emai @quantumwise.com.

thanks very much!