Skip to content

Minor addition of some variables in SetupFromSpglib()#122

Merged
mesonepigreco merged 2 commits into
SSCHAcode:masterfrom
ManexAlkorta:master
Jul 20, 2026
Merged

Minor addition of some variables in SetupFromSpglib()#122
mesonepigreco merged 2 commits into
SSCHAcode:masterfrom
ManexAlkorta:master

Conversation

@ManexAlkorta

Copy link
Copy Markdown
Contributor

Addition to support the new function get_free_energy_hessian_dev() in python-sscha

@ManexAlkorta ManexAlkorta reopened this Jul 2, 2026
@mesonepigreco mesonepigreco self-assigned this Jul 17, 2026
@mesonepigreco

Copy link
Copy Markdown
Collaborator

Hi Manex,
Can you briefly explain what the new variables are used for and what their expected sizes are? This in cellconstructor should be a really trivial addition; however, we must be careful, as they are allocated a lot inside the sscha code, and I see they are a 3-rank tensor (I assume symmetries X nat X 3, am I right? Is the structure in the primitive or in the supercell? In this case, it should be safe for memory, as it is likely the same memory as one configuration of the ensemble per symmetry, and usually we have a much larger number of configurations than symmetries)

@mesonepigreco

Copy link
Copy Markdown
Collaborator

@ManexAlkorta It seems there is a failure in one of the test, can you check?

@ManexAlkorta

Copy link
Copy Markdown
Contributor Author

Hi Lorenzo,

This pull request includes two new new attributes to the QE_symmetry class:

  • self.QE_s_cart. Dimension [3,3,48]. Point group symmetries of the crystal in cartesian coordinates, let's call it S.
  • self.QE_s_inv_cart. Dimension [3,3,48]. Inverse of the point group symmetries of the crystal, S.T or S^{-1}.

In the end these are just the point symmtry operation in cartesian coordinates of the crystal, S and S.T. This allows reading it directly from the SetupFromSpglib function instead of making the conversion from crystal units in python-sscha.

Regarding the conflict in the test, I think it has nothing to do with this commit, but with the master branch that i cloned to add these changes. I checked the master branch (commit 311af57) and i see the same issue:

FAILED tests/ImportFromPhonons/test_asephonons.py::test_ase_phonons - AttributeError: 'Phonons' object has no attribute 'N_c'

@mesonepigreco

Copy link
Copy Markdown
Collaborator

Hi @ManexAlkorta , you were right, it is a new ASE deprecation that made the test not pass when installed with newer versions (a time bomb -.-). I have fixed the error in the master. Can you update the branch to the latest master commit so we can rerun the test suite in CI to verify that no other errors occur?

@ManexAlkorta

Copy link
Copy Markdown
Contributor Author

I updated merged the branch into master in my forked repo. I guess now should be fine

@mesonepigreco
mesonepigreco merged commit 88b8cff into SSCHAcode:master Jul 20, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants