PseudoBands Procedure Help

19 views
Skip to first unread message

Arabi Seshappan

unread,
Apr 2, 2026, 11:21:50 AMApr 2
to BerkeleyGW Help
Dear users and developers,

I am attempting to do a calculation using ParaBands and stochastic pseudobands for the first time, and I'm running into procedural issues.

I first ran a ParaBands calculation, missing the optional keyword "use_pseudobands." It's not clear to me what this keyword actually does--was missing this step a fatal flaw?

Following ParaBands, I used pseudobands.py to create my compressed WFN and WFNq. I then ran an epsilon calculation, which gave me errors regarding the degeneracy and normalization of my states. The documentation shows that if pseudobands are detected, these checks are automatically ignored, so it seems that it doesn't realize I'm using pseudobands.

If I were to proceed with an epsilon calculation using the flags "degeneracy_check_override" and "dont_check_norms" is that enough?

Is it best to proceed with just my ParaBands output as WFN for epsilon, to redo ParaBands with the "use_pseudobands" flag, or to continue with my current pseudobands outputs while adjusting my epsilon flags?

Thanks in advance for any advice!

Best,
Arabi
Reply all
Reply to author
Forward
0 new messages