Apply MP corrections in MP recipes (#1820)

Closes #1747.

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

src/quacc/recipes/vasp/_base.py

@@ -23,6 +23,7 @@ def base_fn(
  preset: str | None = None,
  calc_defaults: dict[str, Any] | None = None,
  calc_swaps: dict[str, Any] | None = None,
+ report_mp_corrections: bool = False,
  additional_fields: dict[str, Any] | None = None,
  copy_files: SourceDirectory | dict[SourceDirectory, Filenames] | None = None,
 ) -> VaspSchema:
@@ -41,6 +42,8 @@ def base_fn(
  Dictionary of custom kwargs for the Vasp calculator. Set a value to
  `None` to remove a pre-existing key entirely. For a list of available
  keys, refer to [quacc.calculators.vasp.vasp.Vasp][].
+ report_mp_corrections
+ Whether to report the Materials Project corrections in the results.
  additional_fields
  Additional fields to supply to the summarizer.
  copy_files
@@ -56,4 +59,8 @@ def base_fn(
  atoms.calc = Vasp(atoms, preset=preset, **calc_flags)
  final_atoms = run_calc(atoms, copy_files=copy_files)
 
- return vasp_summarize_run(final_atoms, additional_fields=additional_fields)
+ return vasp_summarize_run(
+ final_atoms,
+ report_mp_corrections=report_mp_corrections,
+ additional_fields=additional_fields,
+ )

src/quacc/recipes/vasp/mp.py

@@ -74,6 +74,7 @@ def mp_gga_relax_job(
  atoms,
  calc_defaults=calc_defaults,
  calc_swaps=calc_kwargs,
+ report_mp_corrections=True,
  additional_fields={"name": "MP GGA Relax"},
  copy_files=copy_files,
  )
@@ -120,6 +121,7 @@ def mp_gga_static_job(
  atoms,
  calc_defaults=calc_defaults,
  calc_swaps=calc_kwargs,
+ report_mp_corrections=True,
  additional_fields={"name": "MP GGA Static"},
  copy_files=copy_files,
  )
@@ -173,6 +175,7 @@ def mp_metagga_prerelax_job(
  atoms,
  calc_defaults=calc_defaults,
  calc_swaps=calc_kwargs,
+ report_mp_corrections=True,
  additional_fields={"name": "MP Meta-GGA Pre-Relax"},
  copy_files=copy_files,
  )
@@ -222,6 +225,7 @@ def mp_metagga_relax_job(
  atoms,
  calc_defaults=calc_defaults,
  calc_swaps=calc_kwargs,
+ report_mp_corrections=True,
  additional_fields={"name": "MP Meta-GGA Relax"},
  copy_files=copy_files,
  )
@@ -272,6 +276,7 @@ def mp_metagga_static_job(
  atoms,
  calc_defaults=calc_defaults,
  calc_swaps=calc_kwargs,
+ report_mp_corrections=True,
  additional_fields={"name": "MP Meta-GGA Static"},
  copy_files=copy_files,
  )

src/quacc/schemas/vasp.py

@@ -12,6 +12,10 @@
 from monty.os.path import zpath
 from pymatgen.command_line.bader_caller import bader_analysis_from_path
 from pymatgen.command_line.chargemol_caller import ChargemolAnalysis
+from pymatgen.entries.compatibility import (
+ CompatibilityError,
+ MaterialsProject2020Compatibility,
+)
 
 from quacc import SETTINGS
 from quacc.schemas.ase import summarize_run
@@ -36,6 +40,7 @@ def vasp_summarize_run(
  run_bader: bool | None = None,
  run_chargemol: bool | None = None,
  check_convergence: bool | None = None,
+ report_mp_corrections: bool = False,
  additional_fields: dict[str, Any] | None = None,
  store: Store | None = None,
 ) -> VaspSchema:
@@ -62,6 +67,8 @@ def vasp_summarize_run(
  check_convergence
  Whether to throw an error if convergence is not reached. Defaults to True in
  settings.
+ report_mp_corrections
+ Whether to apply the MP corrections to the task document. Defaults to False.
  additional_fields
  Additional fields to add to the task document.
  store
@@ -85,7 +92,21 @@ def vasp_summarize_run(
 
  # Fetch all tabulated results from VASP outputs files. Fortunately, emmet
  # already has a handy function for this
- vasp_task_doc = TaskDoc.from_directory(dir_path).model_dump()
+ vasp_task_model = TaskDoc.from_directory(dir_path)
+
+ # Get MP corrections
+ if report_mp_corrections:
+ mp_compat = MaterialsProject2020Compatibility()
+ try:
+ corrected_entry = mp_compat.process_entry(
+ vasp_task_model.structure_entry, on_error="raise"
+ )
+ vasp_task_model.entry = corrected_entry
+ except CompatibilityError as err:
+ logger.warning(err)
+
+ # Convert the VASP task model to a dictionary
+ vasp_task_doc = vasp_task_model.model_dump()
 
  # Check for calculation convergence
  if check_convergence and vasp_task_doc["state"] != "successful":

tests/core/atoms/test_atoms_core.py

@@ -3,7 +3,6 @@
 import numpy as np
 import pytest
 from ase import Atoms
-from ase.atoms import Atoms
 from ase.build import bulk, molecule
 from ase.io import read
 

tests/core/recipes/espresso_recipes/conftest.py

@@ -1,7 +1,7 @@
 import pytest
 
 
-@pytest.fixture
+@pytest.fixture()
 def ESPRESSO_PARALLEL_INFO():
  from shutil import which
 

tests/core/recipes/espresso_recipes/test_phonons.py

@@ -59,7 +59,9 @@ def test_phonon_job(tmp_path, monkeypatch, ESPRESSO_PARALLEL_INFO):
  )
 
  ph_results = phonon_job(
- pw_results["dir_name"], input_data=ph_loose, parallel_info=ESPRESSO_PARALLEL_INFO,
+ pw_results["dir_name"],
+ input_data=ph_loose,
+ parallel_info=ESPRESSO_PARALLEL_INFO,
  )
 
  assert_allclose(
@@ -171,7 +173,9 @@ def test_q2r_job(tmp_path, monkeypatch, ESPRESSO_PARALLEL_INFO):
  additional_cards = ["1 1 1", "1", "matdyn"]
 
  q2r_results = q2r_job(
- tmp_path, additional_cards=additional_cards, parallel_info=ESPRESSO_PARALLEL_INFO
+ tmp_path,
+ additional_cards=additional_cards,
+ parallel_info=ESPRESSO_PARALLEL_INFO,
  )
 
  assert Path(q2r_results["dir_name"], "q2r.fc.gz").exists()

tests/core/recipes/gulp_recipes/conftest.py

@@ -16,6 +16,7 @@ def mock_settings_env_vars():
 
 def mock_execute(self, *args, **kwargs):
  from shutil import copy
+
  copy(GULP_DIR / "gulp.got.gz", Path(self.directory, "gulp.got.gz"))
  copy(GULP_DIR / "gulp.cif.gz", Path(self.directory, "gulp.cif.gz"))
  copy(GULP_DIR / "gulp.xyz.gz", Path(self.directory, "gulp.xyz.gz"))
@@ -31,6 +32,7 @@ def patch_execute(monkeypatch):
 def mock_read_results(self, *args, **kwargs):
  from ase.calculators.lj import LennardJones
  from ase.io import read
+
  atoms = read(Path(self.directory, "gulp.xyz.gz"))
  atoms.calc = LennardJones()
  atoms.get_potential_energy()

tests/core/recipes/vasp_recipes/mocked/conftest.py

@@ -54,5 +54,4 @@ def mock_taskdoc(*args, **kwargs):
 
 @pytest.fixture(autouse=True)
 def patch_taskdoc(monkeypatch):
-
  monkeypatch.setattr("quacc.schemas.vasp.TaskDoc.from_directory", mock_taskdoc)

tests/core/recipes/vasp_recipes/mocked/test_vasp_recipes.py

@@ -13,9 +13,8 @@
  mp_metagga_static_job,
 )
 from quacc.recipes.vasp.qmof import qmof_relax_job
-from quacc.recipes.vasp.slabs import bulk_to_slabs_flow
+from quacc.recipes.vasp.slabs import bulk_to_slabs_flow, slab_to_ads_flow
 from quacc.recipes.vasp.slabs import relax_job as slab_relax_job
-from quacc.recipes.vasp.slabs import slab_to_ads_flow
 from quacc.recipes.vasp.slabs import static_job as slab_static_job
 
 DEFAULT_SETTINGS = SETTINGS.model_copy()

tests/core/schemas/test_ase.py

@@ -22,7 +22,7 @@
 
 FILE_DIR = Path(__file__).parent
 
-RUN1 = FILE_DIR / "vasp_run1"
+RUN1 = FILE_DIR / "test_files" / "vasp_run1"
 
 
 def test_summarize_run():

tests/core/schemas/test_cclib_schema.py

@@ -23,9 +23,9 @@
 
 FILE_DIR = Path(__file__).parent
 
-run1 = FILE_DIR / "gaussian_run1"
+run1 = FILE_DIR / "test_files" / "gaussian_run1"
 log1 = run1 / "Gaussian.log"
-run2 = FILE_DIR / "cclib_data"
+run2 = FILE_DIR / "test_files" / "cclib_data"
 log2 = run2 / "gau_testopt.log.gz"
 
 
@@ -35,7 +35,7 @@ def cclib_obj():
 
 
 def setup_module():
- p = FILE_DIR / "cclib_data"
+ p = FILE_DIR / "test_files" / "cclib_data"
 
  with (
  gzip.open(p / "psi_test.cube.gz", "r") as f_in,
@@ -45,7 +45,7 @@ def setup_module():
 
 
 def teardown_module():
- p = FILE_DIR / "cclib_data"
+ p = FILE_DIR / "test_files" / "cclib_data"
 
  if os.path.exists(p / "psi_test.cube"):
  os.remove(p / "psi_test.cube")
@@ -143,7 +143,7 @@ def test_errors():
 def test_cclib_taskdoc(tmp_path, monkeypatch):
  monkeypatch.chdir(tmp_path)
 
- p = FILE_DIR / "cclib_data"
+ p = FILE_DIR / "test_files" / "cclib_data"
 
  # Now we will try two possible extensions, but we will make sure that
  # it fails because the newest log file (.txt) is not valid
@@ -196,34 +196,36 @@ def test_cclib_calculate(tmp_path, monkeypatch, cclib_obj):
  _cclib_calculate(
  cclib_obj,
  method="ddec6",
- cube_file=FILE_DIR / "cclib_data" / "psi_test.cube",
+ cube_file=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
  proatom_dir="does_not_exists",
  )
 
  with pytest.raises(OSError):
  _cclib_calculate(
  cclib_obj,
  method="ddec6",
- cube_file=FILE_DIR / "cclib_data" / "psi_test.cube",
+ cube_file=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
  )
 
  with pytest.raises(Exception):
  _cclib_calculate(
  cclib_obj,
  method="ddec6",
- cube_file=FILE_DIR / "cclib_data" / "psi_test.cube",
- proatom_dir=FILE_DIR / "cclib_data" / "psi_test.cube",
+ cube_file=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
+ proatom_dir=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
  )
 
 
 def test_monkeypatches(tmp_path, monkeypatch, cclib_obj, caplog):
  monkeypatch.chdir(tmp_path)
- monkeypatch.setenv("PROATOM_DIR", str(FILE_DIR / "cclib_data" / "proatomdata"))
+ monkeypatch.setenv(
+ "PROATOM_DIR", str(FILE_DIR / "test_files" / "cclib_data" / "proatomdata")
+ )
  with pytest.raises(FileNotFoundError):
  _cclib_calculate(
  cclib_obj,
  method="ddec6",
- cube_file=FILE_DIR / "cclib_data" / "psi_test.cube",
+ cube_file=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
  )
 
  monkeypatch.setattr("cclib.method.Bader.calculate", bad_mock_cclib_calculate)
@@ -232,7 +234,7 @@ def test_monkeypatches(tmp_path, monkeypatch, cclib_obj, caplog):
  _cclib_calculate(
  cclib_obj,
  method="bader",
- cube_file=FILE_DIR / "cclib_data" / "psi_test.cube",
+ cube_file=FILE_DIR / "test_files" / "cclib_data" / "psi_test.cube",
  )
  is None
  )

tests/core/schemas/test_vasp_schema.py

@@ -17,11 +17,16 @@
 
 @pytest.fixture()
 def run1():
- from pathlib import Path
+ FILE_DIR = Path(__file__).parent
+
+ return FILE_DIR / "test_files" / "vasp_run1"
 
+
+@pytest.fixture()
+def mp_run1():
  FILE_DIR = Path(__file__).parent
 
- return FILE_DIR / "vasp_run1"
+ return FILE_DIR / "test_files" / "vasp_mp_run1"
 
 
 def mock_bader_analysis(*args, **kwargs):
@@ -196,6 +201,25 @@ def test_summarize_bader_and_chargemol_run(monkeypatch, run1, tmp_path):
  assert results["bader"]["spin_moments"] == [0.0] * len(atoms)
 
 
+def test_summarize_mp(monkeypatch, mp_run1, tmp_path):
+ monkeypatch.chdir(tmp_path)
+ p = tmp_path / "vasp_run"
+ copytree(mp_run1, p)
+ atoms = read(p / "OUTCAR.gz")
+ results = vasp_summarize_run(atoms, dir_path=p, report_mp_corrections=True)
+ assert results["entry"].correction == pytest.approx(-3.2279999999999998)
+
+
+def test_summarize_mp_bad(monkeypatch, run1, tmp_path, caplog):
+ monkeypatch.chdir(tmp_path)
+ p = tmp_path / "vasp_run"
+ copytree(run1, p)
+ atoms = read(p / "OUTCAR.gz")
+ with caplog.at_level(logging.WARNING):
+ vasp_summarize_run(atoms, dir_path=p, report_mp_corrections=True)
+ assert "invalid run type" in caplog.text
+
+
 def test_no_bader(tmp_path, monkeypatch, run1, caplog):
  monkeypatch.chdir(tmp_path)
 

tests/core/settings/test_settings.py

@@ -46,7 +46,7 @@ def test_results_dir(tmp_path, monkeypatch):
 def test_env_var(monkeypatch, tmp_path):
  p = tmp_path / "my/scratch/dir"
  monkeypatch.setenv("QUACC_SCRATCH_DIR", p)
- assert QuaccSettings().SCRATCH_DIR == p.expanduser().resolve()
+ assert p.expanduser().resolve() == QuaccSettings().SCRATCH_DIR
 
 
 def test_env_var2(monkeypatch, tmp_path):
@@ -80,4 +80,4 @@ def test_yaml(tmp_path, monkeypatch):
  with open(tmp_path / "quacc_test.yaml", "w") as f:
  f.write(f"SCRATCH_DIR: {p}")
  monkeypatch.setenv("QUACC_CONFIG_FILE", os.path.join(tmp_path, "quacc_test.yaml"))
- assert QuaccSettings().SCRATCH_DIR == p.expanduser().resolve()
+ assert p.expanduser().resolve() == QuaccSettings().SCRATCH_DIR