MPMorph Flows (#938)

* Add NEP MLIP relax, static, and MD makers with tests

* Add `SevenNetRelaxMaker` + `SevenNetStaticMaker` to force field jobs

* Add elastic workflow and equation of state workflow for FHI-aims

* Add possibility to use custom M3GNet potentials instead of only pretrained models

* Add new documentation:
  - Document Models / emmet tutorial
  - High level overview of atomate2 concepts
  - Tutorial for blob storage

* Fix CP2K TaskDocument schema

* Add additional fields as kwargs to PhononBSDOSDoc

* Fix JobStoreDocument attribute access

* Allow bulk supercell calculation to be skipped in defect workflow

* Clean up and standardize MPMorph workflows:
  - Refactor MPMorph makers for both VASP and force fields
  - Add from_temperature_and_steps classmethod
  - Improve equilibration output consistency
  - Add tests for all MPMorph workflows

* Various dependency updates and maintenance:
  - Update Python package dependencies
  - Fix linting issues
  - Clean up documentation

---------

Co-authored-by: Aaron Kaplan <[email protected]>
Co-authored-by: J. George <[email protected]>
Co-authored-by: Janosh Riebesell <[email protected]>

Author: 4 people

pyproject.toml

@@ -54,6 +54,7 @@ forcefields = [
     "calorine<=2.2.1",
     "chgnet>=0.2.2",
     "mace-torch>=0.3.3",
+    "torchdata<=0.7.1",
     "matgl>=1.1.3",
     # quippy-ase support for py3.12 tracked in https://github.com/libAtoms/QUIP/issues/645
     "quippy-ase>=0.9.14; python_version < '3.12'",

src/atomate2/ase/jobs.py

@@ -3,6 +3,7 @@
 from __future__ import annotations
 
 import logging
+from abc import ABCMeta, abstractmethod
 from dataclasses import dataclass, field
 from typing import TYPE_CHECKING
 
@@ -28,7 +29,7 @@
 
 
 @dataclass
-class AseMaker(Maker):
+class AseMaker(Maker, metaclass=ABCMeta):
     """
     Define basic template of ASE-based jobs.
 
@@ -44,8 +45,6 @@ class AseMaker(Maker):
         The name of the job
     calculator_kwargs : dict
         Keyword arguments that will get passed to the ASE calculator.
-    calculator_kwargs : dict
-        Keyword arguments that will get passed to the ASE calculator.
     ionic_step_data : tuple[str,...] or None
         Quantities to store in the TaskDocument ionic_steps.
         Possible options are "struct_or_mol", "energy",
@@ -70,6 +69,7 @@ class AseMaker(Maker):
     store_trajectory: StoreTrajectoryOption = StoreTrajectoryOption.NO
     tags: list[str] | None = None
 
+    @abstractmethod
     def run_ase(
         self,
         mol_or_struct: Structure | Molecule,
@@ -92,6 +92,7 @@ def run_ase(
         raise NotImplementedError
 
     @property
+    @abstractmethod
     def calculator(self) -> Calculator:
         """ASE calculator, method to be implemented in subclasses."""
         raise NotImplementedError

src/atomate2/ase/md.py

@@ -7,6 +7,7 @@
 import os
 import sys
 import time
+from abc import ABCMeta, abstractmethod
 from collections.abc import Sequence
 from dataclasses import dataclass, field
 from enum import Enum
@@ -78,7 +79,7 @@ class DynamicsPresets(Enum):
 
 
 @dataclass
-class AseMDMaker(AseMaker):
+class AseMDMaker(AseMaker, metaclass=ABCMeta):
     """
     Perform MD with the Atomic Simulation Environment (ASE).
 
@@ -393,6 +394,7 @@ def _callback(dyn: MolecularDynamics = md_runner) -> None:
         )
 
     @property
+    @abstractmethod
     def calculator(self) -> Calculator:
         """ASE calculator, to be overwritten by user."""
         raise NotImplementedError