Adding then supercell estimator in the Settings Panel. #88
Conversation
Limitations: - use only the structure and the supercell size, not the other parameters (symprec, distinguish_kinds, is_symmetry) - to understand how we can adapt this to consider also the HubbardStructureData - performance is low (especially for the `_reset_supercell` reaction.
- update the number of supercell only when all the three vectors are updated (from hint) or reset - we provide also spinning loading icon when we compute the number of supercells. - missing: symprec integration.
…uted - adding a reset for symprec - adding upper and lower bound 1 and 1e-7 (if out of this range, automatically reset to max or min) - added tests for all the new widgets/logics
|
Hi @AndresOrtegaGuerrero, I think the PR is ready. for now, we don't consider hubbard in the estimation of the supercells to be simulated, neither the dielectric workchain (i.e. how many finite Electric field calculations). I added also symprec (merged from main) in the estimation logic, and also a spinning circle while we wait for the number of supercells to be determined. Please try it with SiO2 or gold, to see it in action for some seconds. Thanks! |
|
|
||
| return | ||
|
|
||
| def _reset_supercell(self, _=None): |
There was a problem hiding this comment.
does it work with _reset_supercell(self): ?
| def _reset_supercell(self, _=None): | ||
| if self.input_structure is not None: | ||
| reset_supercell = [] | ||
| self.block = True |
There was a problem hiding this comment.
Could you add some dockstirng about this variable self.block
| if self.input_structure is not None: | ||
| reset_supercell = [] | ||
| self.block = True | ||
| for direction, periodic in zip( |
There was a problem hiding this comment.
or direction, periodic in zip([self._sc_x, self._sc_y, self._sc_z], self.input_structure.pbc):
reset_supercell.append(2 if periodic else 1) , , what about this ? or is not that clear the way i suggest?
| if self.block: | ||
| return | ||
|
|
||
| if self.symmetry_symprec.value > 1: |
There was a problem hiding this comment.
self.symmetry_symprec.value = max(1e-5, min(self.symmetry_symprec.value, 1)) ,
one question , this is enforcing symprec to always be 1e-5 , what if the user sets a symprec like 1e-3 ?
This address #34
Limitations:
_reset_supercellreaction.