You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently (as of DFTK v0.6.18) there is an error when one tries to give a custom pseudopotential a new identifier (to avoid confusion with the default parameters) and then tries to extract the irrfbz_path from the model.
using DFTK
functioncreate_silicon_model(psp)
a =10.26# Silicon lattice constant in Bohr
lattice = a /2* [[011.];
[101.];
[110.]]
Si =ElementPsp(:Si; psp)
atoms = [Si, Si]
positions = [ones(3)/8, -ones(3)/8]
model_LDA(lattice, atoms, positions; temperature=1e-2)
end
psp_hgh =load_psp("hgh/lda/si-q4.hgh")
irrfbz_path(create_silicon_model(psp_hgh)) # works as expected
psp_custom =PspHgh{Float64}(4, 0.44, [-7.33610297, 0.0, 0.0, 0.0], 1, [0.42273813, 0.48427842], [[5.90692831-1.26189397; -1.261893973.25819622], [2.72701346;;]], "my_custom_identifier", "my custom name")
irrfbz_path(create_silicon_model(psp_custom)) # throws an error
I think the issue is that in external/spglib.jl:35 we make a. AtomsBase system and this should drop the pseudo identifier. Look at the function atomic_system(::Model) in external/atomsbase.jl. There the issue should be.
Currently (as of DFTK v0.6.18) there is an error when one tries to give a custom pseudopotential a new identifier (to avoid confusion with the default parameters) and then tries to extract the irrfbz_path from the model.
The full stack trace:
The text was updated successfully, but these errors were encountered: