Adding ModelMethod for TB Wannier90 #8

JosePizarro3 · 2024-02-20T08:44:10Z

A brief intro before our meeting tomorrow 27.02.2024.

This is my initial version on defining ModelMethod, some base classes which will be contained within, and some usage in the specific case of TB models.

As you can see:

ModelMethod serves as the base class from which to inherit to specific methodologies (e.g., DFT should also be inheriting from it, as well as GW, BSE, ForceFields(?), etc.).
It also contains info about meshes, self-consistency (this is the former schema Scf class), and proxy references to other ModelMethod sections already existing in the archive.

I deleted some quantities and sections which were not so clear or unused:

Smearing
Electronic
MoleculeParameters
LatticeModelHamiltonian
Method.stress_tensor_method, and I defined methods_ref as a generic way of coverng all method references under a given ModelMethod.
Scf.native_tier

I am still working out the details of TB, but there you can see the usage:

Of referencing specific sections in ModelSystem
Of using this class for further inheritance (there are different types of TB models, Wannier, SlaterKoster, xTB).
If you just check the old schema, you would see how complex SlaterKoster was to define, as Mohammad (who was the one doing this) was developing this it was very complex due to the problem of composing TB under ModelMethod. Now, I'll argue that it works much more nicely and it is easily usable by external people.

Now, I will work a bit further on defining the core methodologies (DFT, GW, BSE, DMFT), and you might wanna take over to define your task-specific problems.

JosePizarro3 · 2024-02-26T09:26:21Z

@ndaelman-hu @JFRudzinski I updated the comment in the description and added you as reviewers. Let's chat tomorrow.

Commented out Pseudopotentials

Improved ModelSystem description

Moved lattice_vectors_reciprocal to KMesh Added normalizations

Deleted unused comment in ModelSystem Clean up descriptions in model_system.py

Transfering more sections to ModelMethod

Added self.name under TB

Added common.py for common sections

Improved code writing for assignements in atoms_state.py

… active_atoms in ModelSystem for TB class Improved SlaterKoster method and added normalization for bond name extraction

Fixed orbitals_ref in TB

Improved schema for those

Improved descriptions in GW, TB, BSE

JosePizarro3 · 2024-02-29T09:22:48Z

@ndaelman-hu @JFRudzinski would you mind reviewing this PR?

You can ignore the changes in atoms_state.py. This is done because of the unit testing showed some errors in that schema that I needed to fix.

Everything else in model_method.py is new. Like we talked, I left out everything related with BasisSet and ForceField. I am also skipping the implementation of some functions in DMFT, because first I need to see how the schema behaves as a whole.

JFRudzinski

This is great! Just to bring my main questions to the forefront:

When should we be using Enum values for method names?
When should we be adding eln annotations to quantities?
What is our overall strategy for referencing between methods and the other main sections (workflows, systems, outputs)?

simulationdataschema/atoms_state.py

simulationdataschema/model_method.py

ndaelman-hu · 2024-02-29T14:33:23Z

@JosePizarro3 I'll do the review this evening, okay?

ndaelman-hu

Hey! Very nice work. Mostly small stuff.

Tbh, I didn't check the methods after TB much, but I can see to help there with the testing of the normalizers.
I will brainstorm a bit on how to improve the DFT section too.

.github/workflows/actions.yaml

simulationdataschema/atoms_state.py

simulationdataschema/model_method.py

Added github action for Ruff check and formatting

JosePizarro3 · 2024-03-01T14:08:20Z

Thanks a lot for your reviews!! 👏🏻

@ndaelman-hu @JFRudzinski would you mind to quickly checking again? Focus on your comments, close them if you feel satisfied, and if everything is good, I will merge.

ndaelman-hu · 2024-03-01T18:07:26Z

Thanks a lot for your reviews!! 👏🏻

@ndaelman-hu @JFRudzinski would you mind to quickly checking again? Focus on your comments, close them if you feel satisfied, and if everything is good, I will merge.

I was just about to head home. Do you mind if I do it tomorrow morning?

ndaelman-hu

Okay, all done! I closed all that is resolved from my side. Just check the open comments.

I'm really eager to get hands-on with this schema, as now we will have much clearer business logic.

simulationdataschema/model_method.py

Added utils is_not_representative Redefine QuasiparticlesFrequencyMesh Changed resolve_points to resolve_points_and_offset in KMesh

JosePizarro3 · 2024-03-04T12:04:14Z

Thanks a lot for the reviews @JFRudzinski @ndaelman-hu !! 🥳

I merge this and work on other minor topics of the repo.

JosePizarro3 self-assigned this Feb 20, 2024

JosePizarro3 linked an issue Feb 20, 2024 that may be closed by this pull request

Adding ModelMethod for TB Wannier90 #3

Closed

JosePizarro3 force-pushed the 3-adding-modelmethod-for-tb-wannier90-2 branch 2 times, most recently from 49a6118 to 04bb96b Compare February 26, 2024 09:16

JosePizarro3 requested review from JFRudzinski and ndaelman-hu February 26, 2024 09:26

JosePizarro3 added 22 commits February 26, 2024 10:32

Adding AtomicState and all the other sub-sections

68fa7f3

Improved descriptions and typing

ebbb068

Defined AtomState, Orbitals, CoreHole and HubbardInteractions

95e6ea0

Commented out Pseudopotentials

Changed atom_state for atoms_state

e68e9a6

Updated for Orbitals schema

1442175

Improved schema, normalization, and descriptions of atoms_state.py

12df48d

Improved typos and fix descriptions

c7da83d

Added plural to OrbitalState

efc89ff

Improved ModelSystem description

Added condition for umn and orbitals_ref in HubbardInteractions

210382b

Added meshes and TB model_method base classes

061c0dd

Moved lattice_vectors_reciprocal to KMesh Added normalizations

Fixed OrbitalsState normalization

a9aabe2

Added typing for functions

4055d84

Fix rebase

d895c09

Deleting lattice_vectors_reciprocal from AtomicCell

48e69ed

Deleted unused comment in ModelSystem Clean up descriptions in model_system.py

Fixed degeneracy in AtomsState

50826b6

Added better normalization for KMesh and polish other Mesh sections

2f7eb9f

Transfering more sections to ModelMethod

Fixed ModelMethod basic normalization

56d44e7

Added self.name under TB

Added check for Photon

fbeb3d9

Added atoms_ref and orbitals_ref to TB

43bb51a

Added common.py for common sections

Fixed OrbitalState.__init__

39be162

Improved code writing for assignements in atoms_state.py

Added normalization for referencing AtomsState and OrbitalsState from…

5630bcd

… active_atoms in ModelSystem for TB class Improved SlaterKoster method and added normalization for bond name extraction

Improved SlaterKoster sections

b398c96

JosePizarro3 force-pushed the 3-adding-modelmethod-for-tb-wannier90-2 branch from 04bb96b to b398c96 Compare February 26, 2024 09:32

JosePizarro3 added 8 commits February 27, 2024 12:50

Added NumericalSettings

7c2760a

Fixed orbitals_ref in TB

Fixed OrbitalsState normalization

ac76213

Added DFT and XCFunctional classes

adc1ffc

Fix descriptions

1214588

Added ExcitedStateMethodologies: GW, BSE, Screening, CoreHoleSpectra

9b7ee56

Improved schema for those

Improved descriptions

4417419

Added DMFT

9b04e2b

Improved descriptions in GW, TB, BSE

Align pipeline and toml with simulation workflow plugin

b32d1c4

JFRudzinski approved these changes Feb 29, 2024

View reviewed changes

ndaelman-hu reviewed Mar 1, 2024

View reviewed changes

JosePizarro3 added 2 commits March 1, 2024 14:14

Improved schema from comments by Nathan and Joseph

c9438c5

Added github action for Ruff check and formatting

Fix ruff format

47a471c

JosePizarro3 force-pushed the 3-adding-modelmethod-for-tb-wannier90-2 branch from 08fa77c to 47a471c Compare March 1, 2024 13:41

JosePizarro3 added 2 commits March 1, 2024 14:49

Add verbose to ruff format

2595301

Fix version of ruff

df210d5

JosePizarro3 force-pushed the 3-adding-modelmethod-for-tb-wannier90-2 branch from d8054ce to df210d5 Compare March 1, 2024 14:05

Clean up actions.yaml workflow

331f121

ndaelman-hu approved these changes Mar 2, 2024

View reviewed changes

simulationdataschema/model_method.py Outdated Show resolved Hide resolved

simulationdataschema/model_method.py Outdated Show resolved Hide resolved

JosePizarro3 added 3 commits March 4, 2024 11:45

Added more normalization for LinePathSegment

c032851

Added utils is_not_representative Redefine QuasiparticlesFrequencyMesh Changed resolve_points to resolve_points_and_offset in KMesh

Fix description in multiplicities and weights in KMesh

5ae9c38

Added center, sampling and quadrature to Mesh

4565dc3

JosePizarro3 merged commit 9193c24 into develop Mar 4, 2024
3 checks passed

JosePizarro3 deleted the 3-adding-modelmethod-for-tb-wannier90-2 branch March 4, 2024 12:08

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adding ModelMethod for TB Wannier90 #8

Adding ModelMethod for TB Wannier90 #8

JosePizarro3 commented Feb 20, 2024 •

edited

Loading

JosePizarro3 commented Feb 26, 2024

JosePizarro3 commented Feb 29, 2024

JFRudzinski left a comment

ndaelman-hu commented Feb 29, 2024

ndaelman-hu left a comment

JosePizarro3 commented Mar 1, 2024

ndaelman-hu commented Mar 1, 2024

ndaelman-hu left a comment

JosePizarro3 commented Mar 4, 2024

Adding ModelMethod for TB Wannier90 #8

Adding ModelMethod for TB Wannier90 #8

Conversation

JosePizarro3 commented Feb 20, 2024 • edited Loading

JosePizarro3 commented Feb 26, 2024

JosePizarro3 commented Feb 29, 2024

JFRudzinski left a comment

Choose a reason for hiding this comment

ndaelman-hu commented Feb 29, 2024

ndaelman-hu left a comment

Choose a reason for hiding this comment

JosePizarro3 commented Mar 1, 2024

ndaelman-hu commented Mar 1, 2024

ndaelman-hu left a comment

Choose a reason for hiding this comment

JosePizarro3 commented Mar 4, 2024

JosePizarro3 commented Feb 20, 2024 •

edited

Loading