-
Notifications
You must be signed in to change notification settings - Fork 89
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Tweak bookkeeping in LOBPCG #980
base: master
Are you sure you want to change the base?
Conversation
No, unfortunately not, here is a reproducer (you'll need the Artifacts.toml from dftk in your active project). using LazyArtifacts
using DFTK
using AtomsIO
using Random
lattice = [0.0133767 3.95139 3.81366;
3.70274 -0.113618 3.76523;
3.78117 3.8285 0.0147007]
atoms = [ElementPsp(:Al; psp=load_psp(artifact"pd_nc_sr_pbe_stringent_0.4.1_upf/Al.upf"))]
positions = [zeros(3)]
model = model_PBE(lattice, atoms, positions;
smearing=Smearing.Gaussian(), temperature=1e-3)
basis = PlaneWaveBasis(model; Ecut=10, kgrid=[6,6,6]);
Random.seed!(20)
scfres = self_consistent_field(basis; maxiter=2); |
@antoine-levitt ping |
Hrmf it's still super annoying, I'll give it a shot but no promises...
Sorry I'm dense, but how do I do that? activate DFTK doesn't seem to do |
If you activate / instantiate the lobpcg_fix branch and have AtomsIO installed in your general environment, it should work, since the DFTK source tree already includes the using Pkg; Pkg.add(["AtomsIO"])
Pkg.activate("/path/to/your/location/of/lobpcg_fix/branch")
... paste code above ... |
Updated reproducer with current master: using DFTK
using AtomsIO
using Random
using PseudoPotentialData
lattice = [0.0133767 3.95139 3.81366;
3.70274 -0.113618 3.76523;
3.78117 3.8285 0.0147007]
atoms = [ElementPsp(:Al, PseudoFamily("dojo.nc.sr.pbe.v0_4_1.oncvpsp3.stringent.upf"))]
positions = [zeros(3)]
model = model_DFT(lattice, atoms, positions; functionals=LDA(), smearing=Smearing.Gaussian(), temperature=1e-3)
basis = PlaneWaveBasis(model; Ecut=10, kgrid=[6,6,6])
Random.seed!(20)
scfres = self_consistent_field(basis; maxiter=2) |
@mfherbst does that fix #979?