Read WAVECAR file

Message

oamnehc · #1 Post by **oamnehc** » Fri May 21, 2010 6:12 am

I ran a job for a non-collinear material with vasp4.6, and would like to read the spinor from WAVECAR, for further calculations.
With ISPIN=1 the WAVECAR file lists for each kpt, each planewave, the coefficient for each band (exactly in this order). I now need to confirm (1) how the WAVECAR file lists the coefficients for SPIN=2 cases and (2) how the WAVECAR file lists the spinor information for non-collinear cases.
Thanks for your answer.

panda · #2 Post by **panda** » Wed May 26, 2010 2:33 am

Are you asking how VASP reads it into your continuation calculation or are you actually wanting to convert the WAVECAR from binary to some human readable format?

oamnehc · #3 Post by **oamnehc** » Fri Jun 04, 2010 3:49 am

I'm converting WAVECAR. I have done this for SPIN=1 cases, but now am having difficulty with non-collinear cases.

<span class='smallblacktext'>[ Edited Fri Jun 04 2010, 05:51AM ]</span>

panda · #4 Post by **panda** » Mon Jun 07, 2010 4:45 pm

addressed here:

http://cms.mpi.univie.ac.at/vasp-forum/ ... php?4.7368

oamnehc · #5 Post by **oamnehc** » Thu Jun 10, 2010 8:49 am

Ah-- Good idea to just convert the binary to ASKII, no matter what.

Embarrassed to say that I've never learned python, and do not know well how to do this in any language (if not knowing the layout of the binary file in advance), so would stick to your python script for the time being, and get some help.

Still can't make it work for the time being. The indentation of your script is not correctly shown in this forum page, is it? One more problem, it's better printed to a file, the WAVECAR is GB-huge.

Thanks so much for your info and possibly further help.

panda · #6 Post by **panda** » Wed Jun 16, 2010 8:42 pm

yes, it will be huge - its a huge file. Printing to a file is always an option, suppressing printing of any information you do not deem valuable is another, but then again we have to figure out what information is and is not valuable before we do that!

I just copied the code directly from my IDE - I guess it didn't copy indentation. Sorry, I was under the impression that everyone just knew where to indent regardless. Here is some info about how to indent in python:

http://www.secnetix.de/olli/Python/bloc ... ation.hawk

http://diveintopython.org/getting_to_kn ... _code.html

If you don't like indenting don't chose python, lol!

If you don't know how to do this in any language then I say python is as good as any, especially if you want me to help you, because its what I like using

I should really look over this again, was just playing around with it one day while I was bored and just stopped. If you would like to work on it with me just send me a PM.

luis_vasp · #7 Post by **luis_vasp** » Wed May 18, 2011 10:19 pm

[quote author=1 the WAVECAR file lists for each kpt, each planewave, the coefficient for each band (exactly in this order). I now need to confirm (1) how the WAVECAR file lists the coefficients for SPIN=2 cases and (2) how the WAVECAR file lists the spinor information for non-collinear cases.
Thanks for your answer.[/quote]

luis_vasp · #8 Post by **luis_vasp** » Wed May 18, 2011 10:21 pm

Hi oamnehc, did you ever figure out the format of the WAVECAR binary for noncollinear, spin-polarized calculations. I'll appreciate some help.

Luis

KGov · #9 Post by **KGov** » Tue Feb 18, 2014 6:40 pm

Hello,
Maybe by today someone has figured out the format of WAVECAR for noncollinear calculations?
This really helps: http://www.andrew.cmu.edu/user/feenstra/wavetrans/
but I can't get it working for noncollinear calculations.
Thanks for any help!