Wannier 90 interface in 4.5.2

Anything regarding the post-processing utility (e.g. excitonic wavefunction analysis) is dealt with in this forum.

Moderators: Davide Sangalli, andrea marini, Daniele Varsano

Post Reply
pangrt
Posts: 19
Joined: Thu Jun 18, 2020 1:38 pm

Wannier 90 interface in 4.5.2

Post by pangrt » Fri Apr 02, 2021 4:25 am

Dear all
I try to calculate G0W0 bands structure interpolation according to the example 23 of the wannier90 manual. After using ./ypp,
<---> QP @ K 1 - 102 : b 69 - 76
<---> - S/N 007050 -------------------------- v.04.05.02 r.09122 -
<---> QP db read, writing QP corrections in the .eig Wannier90 format
<---> gw.unsorted.eig written
<---> [07] Memory Overview
<---> [08] Game Over & Game summary
An error occurred when using python3 gw2wannier90.py sese mmn amn
python3 gw2wannier90.py sese mmn amn
------------------------------
##############################
### gw2wannier90 interface ###
##############################
Kpoints number: 1024
Exclude bands option is used: be careful to be consistent with the choice of bands for the GW QP corrections.
G0W0 QP corrections read from sese.gw.unsorted.eig
Adding GW QP corrections to KS eigenvalues
Sorting
GW eigenvalues sorted
------------------------------
writing sese.gw.eig
sese.gw.eig written.
------------------------------

----------
AMN module
----------
Created on 2Apr2021 at 5:57: 1 with SCDM
Traceback (most recent call last):
File "gw2wannier90.py", line 208, in <module>
nb, nk, npr = np.array(s.split(), dtype=int)
ValueError: invalid literal for int() with base 10: '35.000000'.
How can i solve this error?
Thanks!
Pangrt
RongTianPang, Inner Mongolia University, Hohhot, China :)

AntimoMarrazzo
Posts: 18
Joined: Fri Jan 22, 2016 2:46 pm

Re: Wannier 90 interface in 4.5.2

Post by AntimoMarrazzo » Fri Apr 02, 2021 5:23 pm

Dear Pangrt,

could you upload all the input and output files?
Also it would help if you could specify the version of Wannier90 that you are using.
Thanks.

Cheers,
Antimo
Antimo Marrazzo
University of Trieste
Physics Department
Strada Costiera 11, 34151 Trieste (Italy)

pangrt
Posts: 19
Joined: Thu Jun 18, 2020 1:38 pm

Re: Wannier 90 interface in 4.5.2

Post by pangrt » Sat Apr 03, 2021 1:25 pm

Dear Antimo,
Thank you for your reply.I am using wannier-3.1.0.
I found out that the reason for the error was that I used <auto_projections = .true.> instead of
<Begin Projections
C1:sp2;pz
C2:pz
End Projections>. This caused gw2wannier90.py to not run properly.
Thanks!
Pangrt
RongTianPang, Inner Mongolia University, Hohhot, China :)

DavidPolito93
Posts: 51
Joined: Sat Jun 06, 2020 10:43 am

Re: Wannier 90 interface in 4.5.2

Post by DavidPolito93 » Tue Apr 27, 2021 3:02 pm

Dear all,

I make use of this thread since my question is linked to the same topic.
I am using yambo-5.0 and I have performed self-consistent GW computation, so I went beyond the simple G0W0 one.

Is it possible to follow the same procedure described in the wannier90 tutorials for the silicon G0W0 bandstructure in order to obtain the wannierization of the self consistent GW band structure?

Best,

Davide Romanin
-----------------------------------------------------
Post-doctoral fellow
Institut des NanoSciences de Paris
Sorbonne Université, CNRS
4 place Jussieu,
75252 PARIS Cedex 05

AntimoMarrazzo
Posts: 18
Joined: Fri Jan 22, 2016 2:46 pm

Re: Wannier 90 interface in 4.5.2

Post by AntimoMarrazzo » Tue Apr 27, 2021 3:14 pm

DavidPolito93 wrote:
Tue Apr 27, 2021 3:02 pm
I am using yambo-5.0 and I have performed self-consistent GW computation, so I went beyond the simple G0W0 one.
Is it possible to follow the same procedure described in the wannier90 tutorials for the silicon G0W0 bandstructure in order to obtain the wannierization of the self consistent GW band structure?
In principle, the Wannier interface only support G0W0 calculations, being based on the assumption that the Green's function is built using the DFT-KS orbitals, or in other words that the wavefunction is still the DFT one.
However, if you perform the self-consistent cycle on the eigenvalues only, then you can still use the Wannier90 interface, exactly because the wave function is kept at the DFT level and you are not changing the orbitals in the Green's function but only the eigenvalues.

To my knowledge Yambo only supports the self-consistence cycle on the eigenvalues, so you can probably use the Wannier interface as it is :-)

HTH,
Antimo
Antimo Marrazzo
University of Trieste
Physics Department
Strada Costiera 11, 34151 Trieste (Italy)

User avatar
Daniele Varsano
Posts: 2997
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: Wannier 90 interface in 4.5.2

Post by Daniele Varsano » Wed Apr 28, 2021 12:16 pm

Dear David, Antimo,

just a comment to say that from yambo-5.0, the yambo_sc which allow to perform self-consistent calculation in COHSEX framework has been released,
even if it is an experimental feature and the support on that is very limited.

See here the complete release note:
https://github.com/yambo-code/yambo/releases

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

DavidPolito93
Posts: 51
Joined: Sat Jun 06, 2020 10:43 am

Re: Wannier 90 interface in 4.5.2

Post by DavidPolito93 » Wed Apr 28, 2021 12:45 pm

Dear all,

Thank you for your answers!

I'll let you know if I find any problem

Best,

Davide

Post Reply