ReWoTes project from Anup

.github/workflows/python-test.yml

@@ -0,0 +1,32 @@
+# .github/workflows/python-tests.yml
+name: Python package
+
+on: [push, pull_request]
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v2
+
+    - name: Set up Python
+      uses: actions/setup-python@v2
+      with:
+        python-version: '3.10'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install -r requirements.txt
+
+    - name: Install NWChem
+      run: |
+        sudo apt-get update
+        sudo apt-get install nwchem -y  # Install NWChem
+
+    - name: Run tests
+      run: |
+        pytest kumaranu/tests

basis_set_error_data.csv

@@ -0,0 +1,7 @@
+geometry,STO-3G-error-percent,3-21G-error-percent,6-31G-error-percent,6-31G*-error-percent,6-31G**-error-percent
+"[[0.0, 0.0, 0.559859335422516], [0.0, 0.0, -0.559859335422516]]",-1.2652515155542783,-0.4917908825196436,-0.03702286490532115,-0.08503021510053684,-0.08503021510053684
+"[[-0.214321196079254, -0.084094531834126, 0.069681838154793], [1.531570315361023, 0.790129363536835, 0.035592794418335], [-1.336195707321167, 0.05733397603035, -0.031842533499002], [1.138465881347656, -0.085862502455711, -0.028848262503743]]",-1.267831657961443,-0.4727418843765934,-0.04650275463400509,-0.08341676419349328,-0.08694225426471786
+"[[0.014473460614681, 0.497637867927551, -0.0], [-0.188154995441437, 1.678797006607056, -0.0], [0.014473460614681, -0.666374921798706, 0.0]]",-1.2017316341566104,-0.44917784243684294,-0.07095387279989016,-0.10394781418823773,-0.10591267886803132
+"[[0.0, -0.0, -0.68626481294632], [-0.0, 0.0, 0.514698624610901]]",-1.266834206595056,-0.475787564741829,-0.043394708420021647,-0.08409217074733753,-0.08409217074733753
+"[[0.509791612625122, 0.026870004832745, -0.023425199091434], [1.010899424552917, 0.925699234008789, 0.045212235301733], [1.415517210960388, -0.839163362979889, 0.0434595271945], [-0.685645818710327, -0.030969487503171, 0.006484928540885]]",-1.2672524312671152,-0.47041342190954255,-0.052250836697970725,-0.0883306127690224,-0.09078009857240983
+"[[0.0, 0.076925121247768, 0.0], [1.179724931716919, -0.162000626325607, -0.0], [-1.179724931716919, 0.104306787252426, 0.0]]",-1.2997750766051088,-0.48029585658507695,-0.03818469310548968,-0.07979115363540501,-0.07979115363540501

kumaranu/README.md

@@ -0,0 +1,87 @@
+# Basis set selector (Chemistry)
+
+> Ideal candidate: scientists skilled in Density Functional Theory and proficient in python.
+
+# Overview
+
+The aim of this task is to create a simple python package that implements automatic basis set selection mechanism for a quantum chemistry engine.
+
+# Requirements
+
+1. automatically find the basis set delivering a particular precision, passed as argument (eg. within 0.01% from reference)
+1. use either experimental data or higher-fidelity modeling results (eg. coupled cluster) as reference data
+1. example properties to converge: HOMO-LUMO gaps, vibrational frequencies
+
+# Expectations
+
+- mine reference data for use during the project
+- correctly find a basis set that satisfies a desired tolerance for a set of 10-100 molecules, starting from H2, as simplest, up to a 10-20-atom ones
+- modular and object-oriented implementation
+- commit early and often - at least once per 24 hours
+
+# Timeline
+
+We leave exact timing to the candidate. Must fit Within 5 days total.
+
+# User story
+
+As a user of this software I can start it passing:
+
+- molecular structure
+- reference datapoint
+- tolerance (precision)
+
+as parameters and get the basis set that satisfies the tolerance criterion.
+
+# Notes
+
+- create an account at exabyte.io and use it for the calculation purposes
+- suggested modeling engine: NWCHEM or SIESTA
+
+## Getting Started
+
+### Clone the Repository:
+
+```
+git clone https://github.com/kumaranu/rewotes.git
+
+cd rewotes
+```
+Create and Activate Conda Environment:
+```
+conda create -n test0 python=3.10
+conda activate test0
+```
+Install the Package:
+```
+pip install -e .
+```
+Run the python script given below:
+```
+import importlib
+from pathlib import Path
+from ase.atoms import Atoms
+from kumaranu.basisSetProvider import BasisSetProvider
+
+# Define project_root
+kumaranu_spec = importlib.util.find_spec('kumaranu')
+project_root = Path(kumaranu_spec.origin).parent.parent
+
+# Define the molecules
+mol = Atoms('CO2', positions=[[0, 0, 0], [1, 1.01, 1], [-1, -1.03, -1]])
+ref = Atoms('CO2', positions=[[0, 0, 0], [1.01, 1, 1], [-1, -1.01, -1]])
+
+# Set the tolerance
+tolerance = 0.5
+
+# Create the BasisSetProvider object
+basisProviderObject = BasisSetProvider(
+    tolerance,
+    files_dir=str(project_root / 'kumaranu/tests/three_molecules'),
+    recalculate_errors=True,
+)
+
+# Get the selected basis set
+selected_basis = basisProviderObject.get_basis_set(mol, ref)
+print(selected_basis)
+```

kumaranu/__init__.py

@@ -0,0 +1,4 @@
+# kumaranu/__init__.py
+
+__version__ = '0.1'
+

kumaranu/basisSetProvider.py

@@ -0,0 +1,87 @@
+from typing import List
+from ase.atoms import Atoms
+from kumaranu.dataCollector import DataCollector
+from kumaranu.basisSetSelector import BasisSetSelector
+from pathlib import Path
+
+
+class BasisSetProvider:
+    """
+    Provides the best basis set for a given molecular structure within a specified tolerance.
+
+    Parameters
+    ----------
+    tolerance : float
+        The tolerance for selecting the basis set.
+    error_data_file : str, optional
+        The path to the CSV file containing error data for basis sets.
+        If not provided, a default path within the project root will be used.
+    files_dir : str, optional
+        The directory containing molecular XYZ files.
+        If not provided, a default directory within the project root will be used.
+    basis_sets : List[str], optional
+        A list of basis sets to be considered.
+        If not provided, a default list of common basis sets will be used.
+    recalculate_errors : bool, optional
+        Whether to recalculate errors and update the error data file.
+        Default is False.
+
+    Methods
+    -------
+    get_basis_set(molecular_structure, reference_datapoint)
+        Returns the best basis set for the given molecular structure within the specified tolerance.
+    """
+    def __init__(
+            self,
+            tolerance: float,
+            error_data_file: str = None,
+            files_dir: str = None,
+            basis_sets: List[str] = None,
+            recalculate_errors: bool = False,
+    ):
+        self.tolerance = tolerance
+        self.files_dir = files_dir if files_dir else f'{Path(__file__).resolve().parents[2]}/kumaranu/tests/molecule_xyz_files'
+        self.error_data_file = error_data_file if error_data_file \
+            else f'{files_dir}/basis_set_error_data.csv'
+        self.basis_sets = basis_sets if basis_sets else [
+            "STO-3G", "3-21G", "6-31G", "6-31G*", "6-31G**",
+            "6-311G", "6-311G*", "6-311G**", "6-311++G**", "6-311++G(2d,2p)",
+        ]
+        self.recalculate_errors = recalculate_errors
+
+    def get_basis_set(
+            self,
+            molecular_structure: Atoms,
+            reference_datapoint: Atoms,
+    ) -> str:
+        """
+        Returns the best basis set for the given molecular structure within the specified tolerance.
+
+        Parameters
+        ----------
+        molecular_structure : Atoms
+            The molecular structure for which to select the basis set.
+        reference_datapoint : Atoms
+            The reference molecular structure to compare against.
+
+        Returns
+        -------
+        str
+            The selected basis set within the specified tolerance.
+        """
+        if self.recalculate_errors:
+            data_collector = DataCollector(
+                self.files_dir,
+                self.basis_sets,
+            )
+            data_collector.collect_and_store_data()
+        error_data, basis_sets = DataCollector.load_error_data(self.error_data_file)
+
+        selector = BasisSetSelector(
+            molecular_structure,
+            reference_datapoint,
+            self.tolerance,
+            error_data,
+            basis_sets,
+        )
+        return selector.select_basis_set()

kumaranu/basisSetSelector.py

@@ -0,0 +1,79 @@
+import numpy as np
+from ase.atoms import Atoms
+
+
+class BasisSetSelector:
+    """
+    Selects the best basis set for a given molecular structure within a specified tolerance.
+
+    Parameters
+    ----------
+    molecular_structure : Atoms
+        The molecular structure for which the basis set needs to be selected.
+    reference_datapoint : Atoms
+        The reference molecular structure used to compare and find the best basis set.
+    tolerance : float
+        The acceptable error tolerance for selecting the basis set.
+    error_data : dict
+        A dictionary containing error percentages for various basis sets.
+    basis_sets : list
+        A list of basis sets to consider.
+
+    Methods
+    -------
+    select_basis_set()
+        Selects the best basis set that satisfies the error tolerance.
+    """
+    def __init__(
+            self,
+            molecular_structure: Atoms,
+            reference_datapoint: Atoms,
+            tolerance: float,
+            error_data: dict,
+            basis_sets: list,
+    ):
+        self.molecular_structure = molecular_structure
+        self.reference_datapoint = reference_datapoint
+        self.tolerance = tolerance
+        self.error_data = error_data
+        self.basis_sets = basis_sets
+
+    def select_basis_set(self):
+        """
+        Selects the best basis set that satisfies the error tolerance.
+
+        This method compares the error percentages of the available basis sets with the specified tolerance.
+        If a basis set meets the tolerance, it is selected. Otherwise, the basis set with the minimum error is selected.
+
+        Returns
+        -------
+        str
+            The name of the selected basis set.
+
+        Raises
+        ------
+        ValueError
+            If the chemical formula of the new geometry does not match the reference.
+        """
+        new_formula = str(self.molecular_structure.symbols)
+        known_formula = str(self.reference_datapoint.symbols)
+
+        if known_formula != new_formula:
+            raise ValueError(
+                f"The chemical formula for the new geometry ({new_formula}) and "
+                f"the reference ({known_formula}) do not match.",
+            )
+
+        errors = np.abs(np.array(self.error_data[str(self.reference_datapoint.symbols)]))
+        below_tolerance = errors <= self.tolerance
+
+        if any(below_tolerance):
+            selected_index = np.argmax(below_tolerance)
+            selected_basis = self.basis_sets[selected_index][:-14]
+            return selected_basis
+        else:
+            best_index = np.argmin(errors)
+            best_basis = self.basis_sets[best_index][:-14]
+            print(f"Warning: No basis set can satisfy the tolerance of {self.tolerance}. "
+                  f"Using the best available basis set, {best_basis}.")
+            return best_basis