Cody Melton
Cody Melton
The case I currently care about is if one of the series gets killed by wallclock time. If I'm understanding you correctly, you need both the random.h5 and the config.h5...
For reference, if you use nexus it automatically sets the appropriate cusps for you. But I agree, this should be done inside qmcpack for those who don't use nexus.
> Cody: Can we revisit the WIP label of this PR and get it merged? This addition is a great achievement -- several people have been able to successfully use...
I also have a separate tool, python based, but similar implementation to ppconvert, which writes to UPF2 and should have the correct labeling. It also does SOC, so it could...
FWIW, I actually trust the cutoff calculations in the python tool more. I have noticed strange behavior from opium when calculating the cutoffs, i.e. blips in the convergence rather than...
The testing of multiple states is something I need to work on.
Hi @zenandrea, there does seem to be an issue, thanks for pointing it out. I was looking at the source code and the potential files. Inside qmcpack, we actually do...
Also, since this is supposed to be a local only potential, the cutoff is irrelevant in that case. The local potential is always splined on the original grid defined in...
I'm working now adding the derivatives testing, and then will add the other APIs I need to get this optimizable
Some of this is addressed in #5356