Refactoring transferibility verification workflow

pyproject.toml

@@ -43,7 +43,7 @@ aiida-sssp-workflow = "aiida_sssp_workflow.cli:cmd_root"
 "sssp_workflow.birch_murnaghan_fit" = "aiida_sssp_workflow.calculations.birch_murnaghan_fit:birch_murnaghan_fit"
 
 [project.entry-points."aiida.workflows"]
-"sssp_workflow.measure_precision" = "aiida_sssp_workflow.workflows.measure.precision:PrecisionMeasureWorkChain"
+"sssp_workflow.measure.transferability" = "aiida_sssp_workflow.workflows.measure.transferability:EOSTransferabilityWorkChain"
 "sssp_workflow.measure_bands" = "aiida_sssp_workflow.workflows.measure.bands:BandsMeasureWorkChain"
 "sssp_workflow.convergence.caching" = "aiida_sssp_workflow.workflows.convergence.caching:_CachingConvergenceWorkChain"
 "sssp_workflow.convergence.eos" = "aiida_sssp_workflow.workflows.convergence.eos:ConvergenceEOSWorkChain"

src/aiida_sssp_workflow/utils/pseudo.py

@@ -1,6 +1,16 @@
 from pydantic import BaseModel
+from importlib import resources
+from enum import Enum
 
 from pseudo_parser.upf_parser import parse
+from aiida.plugins import DataFactory
+
+UpfData = DataFactory("pseudo.upf")
+
+
+class CurateType(Enum):
+    SSSP = "sssp"
+    NC = "nc"
 
 
 class PseudoInfo(BaseModel):
@@ -43,3 +53,86 @@ def parse_std_filename(filename: str, extension: str = "upf") -> PseudoInfo:
         type=type,
         z_valence=int(num_of_valence),
     )
+
+
+def compute_total_nelectrons(configuration, pseudos):
+    """Compute the number of electrons of oxide configurations with pseudos
+
+    This function is limited to only computer the total number of electrons of oxides.
+    """
+    if len(pseudos) != 2:
+        raise ValueError(
+            f"There are {len(pseudos)} != 2 pseudos, we expect for binary oxides."
+        )
+
+    z_O = None
+    z_X = None
+    for e, p in pseudos.items():
+        if e == "O":
+            z_O = p.z_valence
+        else:
+            z_X = p.z_valence
+
+    if z_O is None or z_X is None:
+        raise ValueError(
+            "Either `O` or `X` pseudos not read properly to get number of valence electrons."
+        )
+
+    if configuration == "XO":
+        return z_X + z_O
+
+    elif configuration == "XO2":
+        return z_X + z_O * 2
+
+    elif configuration == "XO3":
+        return z_X + z_O * 3
+
+    elif configuration == "X2O":
+        return z_X * 2 + z_O
+
+    elif configuration == "X2O3":
+        return z_X * 4 + z_O * 6
+
+    elif configuration == "X2O5":
+        return z_X * 4 + z_O * 10
+    else:
+        raise ValueError(
+            f"Cannot compute the number electrons of configuration {configuration}."
+        )
+
+
+def get_pseudo_O(curate_type: CurateType | str = CurateType.SSSP):
+    """Return pseudo of oxygen for oxides"""
+    match curate_type:
+        case CurateType.SSSP:
+            import_path = resources.path(
+                "aiida_sssp_workflow.statics.upf", "O.paw.pbe.z_6.ld1.psl.v0.1.upf"
+            )
+            with import_path as psp_path, open(psp_path, "rb") as stream:
+                pseudo = UpfData(stream)
+            ecutwfc, ecutrho = 70.0, 560.0
+        case CurateType.NC:
+            import_path = resources.path(
+                "aiida_sssp_workflow.statics.upf",
+                "O.nc.pbe.z_6.oncvpsp3.dojo.v0.4.1-std.upf",
+            )
+            with import_path as psp_path, open(psp_path, "rb") as stream:
+                pseudo = UpfData(stream)
+            ecutwfc, ecutrho = 80.0, 320.0
+        case _:
+            raise ValueError(f"Unknown curate_type = {curate_type}")
+
+    return pseudo, ecutwfc, ecutrho
+
+
+# XXX: should do the same as pseudo_O when using for LAN-Nitride band structure calculation.
+# Depend on the target curate library type we use the corresponding N pseudos.
+def get_pseudo_N():
+    """Return pseudo of nitrogen for lanthanide nitrides"""
+    import_path = resources.path(
+        "aiida_sssp_workflow.statics.upf", "N.us.pbe.z_5.ld1.psl.v0.1.upf"
+    )
+    with import_path as psp_path, open(psp_path, "rb") as stream:
+        pseudo_N = UpfData(stream)
+
+    return pseudo_N, 55.0, 330.0

src/aiida_sssp_workflow/utils/structure.py

@@ -13,6 +13,10 @@
 UNARIE_CONFIGURATIONS = ["BCC", "FCC", "SC", "DC"]
 ACWF_CONFIGURATIONS = OXIDE_CONFIGURATIONS + UNARIE_CONFIGURATIONS
 
+VALID_CONFIGURATIONS = (
+    ACWF_CONFIGURATIONS  # FIXME: should also have GS for bands and LANN
+)
+
 
 @calcfunction
 def get_default_configuration(element: orm.Str, property: orm.Str) -> orm.Str:
@@ -40,7 +44,7 @@ def _get_default_configuration(element: str, property: str) -> str:
 
 @calcfunction
 def get_standard_structure(
-    element: orm.Str, configuration=orm.Str
+    element: orm.Str, configuration: orm.Str
 ) -> orm.StructureData:
     try:
         ase_structure = _get_standard_structure(element.value, configuration.value)
@@ -59,7 +63,7 @@ def get_standard_structure(
     return structure
 
 
-def _get_standard_structure(element: str, configuration=str) -> Atoms:
+def _get_standard_structure(element: str, configuration: str) -> Atoms:
     """
     Create an ASE structure from property and configuration and element.
 

src/aiida_sssp_workflow/workflows/common.py

@@ -1,10 +1,6 @@
-import importlib
 from typing import Optional
 
 from aiida import orm
-from aiida.plugins import DataFactory
-
-UpfData = DataFactory("pseudo.upf")
 
 
 def get_extra_parameters_for_lanthanides(element, nbnd) -> dict:
@@ -35,28 +31,6 @@ def get_extra_parameters_for_lanthanides(element, nbnd) -> dict:
     return extra_parameters
 
 
-def get_pseudo_N():
-    """Return pseudo of nitrogen for lanthanide nitrides"""
-    import_path = importlib.resources.path(
-        "aiida_sssp_workflow.statics.upf", "N.us.z_5.ld1.psl.v0.1.upf"
-    )
-    with import_path as psp_path, open(psp_path, "rb") as stream:
-        pseudo_N = UpfData(stream)
-
-    return pseudo_N
-
-
-def get_pseudo_O():
-    """Return pseudo of oxygen for oxides"""
-    import_path = importlib.resources.path(
-        "aiida_sssp_workflow.statics.upf", "O.paw.z_6.ld1.psl.v0.1.upf"
-    )
-    with import_path as psp_path, open(psp_path, "rb") as stream:
-        pseudo_O = UpfData(stream)
-
-    return pseudo_O
-
-
 def clean_workdir(node: orm.CalcJobNode) -> Optional[int]:
     """clean remote workdir of nonmenon calcjob"""
     # I have to do only clean nonmenon calcjob since I regard it as a bug that

src/aiida_sssp_workflow/workflows/convergence/_base.py

@@ -120,7 +120,7 @@ def define(cls, spec):
         )
 
         spec.output(
-            "convergence_report",
+            "report",
             valid_type=orm.Dict,
             required=True,
             help="The output report of convergence verification, it is a dict contains the full information of convergence test, the mapping of cutoffs to the UUID of the evaluation workchain etc.",
@@ -413,7 +413,7 @@ def inspect_convergence(self):
             raise e
         else:
             self.out(
-                "convergence_report",
+                "report",
                 orm.Dict(dict=validated_report.model_dump()).store(),
             )
 

src/aiida_sssp_workflow/workflows/evaluate/_metric.py

@@ -15,7 +15,7 @@
 
 
 class MetricWorkChain(SelfCleanWorkChain):
-    """WorkChain calculate the bands for certain pseudopotential"""
+    """WorkChain running EOS workflow and compute nu/delta values"""
 
     @classmethod
     def define(cls, spec):

src/aiida_sssp_workflow/workflows/measure/__init__.py

@@ -14,16 +14,6 @@
 class _BaseMeasureWorkChain(SelfCleanWorkChain):
     """Base Measure Workchain since bands measure and precision measure share same input ports"""
 
-    # ECUT for oxygen, remember to update this if the oxygen pseudo is changed
-    # Currently the oxygen pseudo is `O.paw.z_6.ld1.psl.v0.1.upf`
-    _O_ECUTWFC = 70.0
-    _O_ECUTRHO = 560.0
-
-    # ECUT for nitrogen, remember to update this if the nitrogen pseudo is changed
-    # Currently the nitrogen pseudo is `N.us.z_5.ld1.psl.v0.1.upf`
-    _N_ECUTWFC = 55.0
-    _N_ECUTRHO = 330.0
-
     @classmethod
     def define(cls, spec):
         """Define the process specification."""
@@ -33,31 +23,35 @@ def define(cls, spec):
                     help='The `pw.x` code use for the `PwCalculation`.')
         spec.input('pseudo', valid_type=UpfData, required=True,
                     help='Pseudopotential to be verified')
-        spec.input('oxygen_pseudo', valid_type=UpfData, required=False)
-        spec.input('oxygen_ecutwfc', valid_type=orm.Float, required=False)
-        spec.input('oxygen_ecutrho', valid_type=orm.Float, required=False)
+        spec.input('oxygen_pseudo', valid_type=UpfData, required=True)
+        spec.input('oxygen_ecutwfc', valid_type=orm.Float, required=True)
+        spec.input('oxygen_ecutrho', valid_type=orm.Float, required=True)
         spec.input('protocol', valid_type=orm.Str, required=True,
                     help='The protocol which define input calculation parameters.')
-        spec.input('wavefunction_cutoff', valid_type=orm.Float, required=True, help='The wavefunction cutoff.')
         spec.input('configurations', valid_type=orm.List, required=False)
+        spec.input('wavefunction_cutoff', valid_type=orm.Float, required=True, help='The wavefunction cutoff.')
         spec.input('charge_density_cutoff', valid_type=orm.Float, required=True, help='The charge density cutoff.')
-        spec.input('options', valid_type=orm.Dict, required=True,
-                    help='Optional `options` to use for the `PwCalculations`.')
-        spec.input('parallelization', valid_type=orm.Dict, required=True,
-                    help='Parallelization options for the `PwCalculations`.')
+        spec.input(
+            "parallelization",
+            valid_type=orm.Dict,
+            required=False,
+            help="The parallelization settings for the `PwCalculation`.",
+        )
+        spec.input(
+            "mpi_options",
+            valid_type=orm.Dict,
+            required=False,
+            help="The MPI options for the `PwCalculation`.",
+        )
 
     def _get_pw_cutoff(
         self, structure: orm.StructureData, ecutwfc: float, ecutrho: float
     ):
         """Get cutoff pair, if strcture contains oxygen or nitrogen, need
         to use the max between pseudo cutoff and the O/N cutoff.
         """
-        if "oxygen_pseudo" in self.inputs:
-            o_ecutwfc = self.inputs.oxygen_ecutwfc.value
-            o_ecutrho = self.inputs.oxygen_ecutrho.value
-        else:
-            o_ecutwfc = self._O_ECUTWFC
-            o_ecutrho = self._O_ECUTRHO
+        o_ecutwfc = self.inputs.oxygen_ecutwfc.value
+        o_ecutrho = self.inputs.oxygen_ecutrho.value
 
         elements = set(structure.get_symbols_set())
         if "O" in elements: