Antoine Levitt
Antoine Levitt
Discussed today with @gkemlin. We should get rid of n_ep_extra and have n_bands take the place of what is currently n_bands+n_ep_extra. What we should do is that self_consistent_field takes as...
There are two parts to this. 1. Using TRS only to speed up FFTs, keeping the data representation the same 2. Using a minimal compressed representation The issue with 1...
I just found out about FourierTools (https://github.com/bionanoimaging/FourierTools.jl) which is pretty cool, eg to take slices, super or sub sample, etc. I don't think we can use it to replace any...
We implement kpoint MPI right now. We can also distribute wrt bands, and G/r vectors. Band is mostly trivial except for the Rayleigh-Ritz/orthogonalization, which needs to be handled specially. The...
eg https://discourse.julialang.org/t/ann-perfchecker-jl-a-small-collection-of-packages-performance-checking-and-plotting-tools/71307
So, if we omit any kinetic or nonlocal terms, we get thomas-fermi like models where the density is a functional of the density only. That's quite fun for educational purposes,...
Eg for direct minimization and newton. Essentially the only thing that you have to change for MPI is that whenever you have something that couples k points (such a density...
See https://github.com/JuliaMolSim/DFTK.jl/pull/513#event-5159480887. Currently only done for Fermi-Dirac, should be done for other smearing types. Probably simplest to not define rules but have the function return 0 at overflows.
Our entropy term is weird. It's not really a function of the input density matrix (psi, occ), it also depends on the eigenvalues (and there's no one-to-one mapping between them...