Port ThermoML parsing, add data entry types

.github/workflows/gh-ci.yaml

@@ -33,7 +33,7 @@ jobs:
         matrix:
           os: [macOS-latest, ubuntu-latest]
           python-version: ["3.11", "3.12"]
-          include-rdkit: [false, true]
+          include-rdkit: [true]
           include-openeye: [false, true]
           exclude:
             - include-rdkit: false

LICENSE-3RD-PARTY

@@ -0,0 +1,23 @@
+=============================== OpenFF Evaluator ===============================
+
+MIT License
+
+Copyright (c) 2019 Open Force Field Consortium
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

README.md

@@ -70,6 +70,8 @@ Contributions are welcome! Please feel free to submit a Pull Request.
 
 This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
 
+This project derives from other projects. See [LICENSE-3RD-PARTY](LICENSE-3RD-PARTY) for details.
+
 ## Authors
 
 - Lily Wang

devtools/conda-envs/dev.yaml

@@ -8,6 +8,7 @@ dependencies:
   # Base dependencies
   - pydantic >=2.0
   - pyyaml
+  - pandas =2
 
   # Core dependencies
   - openff-toolkit >=0.18
@@ -47,6 +48,5 @@ dependencies:
   # Typing
   - mypy
   - types-PyYAML
-
-  - pip:
-    - git+https://github.com/openforcefield/openff-sphinx-theme.git@main
+  - types-requests
+  - types-python-dateutil

devtools/conda-envs/test.yaml

@@ -9,6 +9,7 @@ dependencies:
   # Base dependencies
   - pydantic >=2.0
   - pyyaml
+  - pandas =2
 
   # Core dependencies
   - openff-toolkit
@@ -47,3 +48,5 @@ dependencies:
   # Typing
   - mypy
   - types-PyYAML
+  - types-requests
+  - types-python-dateutil

dimsim/_tests/datasets/test_thermoml.py

@@ -1,38 +1,33 @@
-import numpy as np
+import uuid
+
+import numpy
 import pytest
 from openff.toolkit import Molecule
+from openff.units import Unit
 
 from dimsim._tests.utils import get_test_data_path
-from dimsim.datasets.thermoml import (
-    thermoml_dataset_from_doi,
-    thermoml_dataset_from_xml,
-)
+from dimsim.datasets.thermoml.thermoml import ThermoMLDataSet
 
 
-@pytest.mark.skip(reason="Not implemented yet")
 @pytest.mark.parametrize(
     "filename, expected",
     [
         (
             "single_density.xml",
             {
-                "type": "density",
-                "x_a": 1.0,
-                "x_b": None,
+                "x": [1.0],
                 "temperature": 293.15,
                 "pressure": 1.0,
                 "value": 0.96488,
-                "std": 0.05,
+                "std": 0.00005,
                 "units": "g/mL",
                 "source": "",
             },
         ),
         (
             "single_dhmix.xml",
             {
-                "type": "dhmix",
-                "x_a": 0.219,
-                "x_b": 0.781,
+                "x": [0.219, 0.781],
                 "temperature": 298.15,
                 "pressure": 0.997,
                 "value": 0.03021,
@@ -44,9 +39,7 @@
         (
             "single_dhvap.xml",
             {
-                "type": "dhvap",
-                "x_a": 1.0,
-                "x_b": None,
+                "x": [1.0],
                 "temperature": 298.15,
                 "pressure": None,
                 "value": 10.51625,
@@ -58,10 +51,8 @@
         (
             "single_dielectric.xml",
             {
-                "type": "dielectric_constant",
-                "x_a": 1.0,
-                "x_b": None,
-                "temperature": 298.15,
+                "x": [1.0],
+                "temperature": 293.15,
                 "pressure": 0.997,
                 "value": 11.76,
                 "std": 0.02,
@@ -71,38 +62,54 @@
         ),
     ],
 )
-def test_load_property_types(filename: str, expected: dict):
-    """Test loading a single data type from a ThermoML XML file"""
-    dataset = thermoml_dataset_from_xml(get_test_data_path(f"thermoml/{filename}"))
-    assert len(dataset) == 1
+class TestThermoMLDataset:
+    """Class set up only to make convenient re-use of parametrized test cases."""
+
+    def test_load_property_types(self, filename: str, expected: dict):
+        """Test loading a single data type from a ThermoML XML file"""
+        dataset = ThermoMLDataSet.from_xml(open(get_test_data_path(f"thermoml/{filename}")).read())
+        assert len(dataset) == 1
+
+        entry = next(iter(dataset))
+        assert entry["x"] == expected["x"]
 
-    entry = dataset[0]
-    assert entry["type"] == expected["type"]
-    assert entry["x_a"] == expected["x_a"]
+        assert len(entry["x"]) == len(entry["smiles"])
+        assert len(entry["x"]) == len(expected["x"])
 
-    # assert mapped smiles
-    Molecule.from_mapped_smiles(entry["smiles_a"])
-    if expected["x_b"] is not None:
-        assert entry["x_b"] == expected["x_b"]
-        Molecule.from_mapped_smiles(entry["smiles_b"])
-    else:
-        assert entry["smiles_b"] is None
-        assert entry["x_b"] is None
+        for found_x, expected_x, found_smiles in zip(
+            entry["x"],
+            expected["x"],
+            entry["smiles"],
+        ):
+            assert found_x == expected_x
 
-    assert entry["temperature"] == expected["temperature"]
-    if expected["pressure"] is not None:
-        assert np.isclose(entry["pressure"], expected["pressure"], atol=1e-3)
-    else:
-        assert entry["pressure"] is None
+            # just make sure it's valid SMILES
+            Molecule.from_smiles(found_smiles)
 
-    assert np.isclose(entry["value"], expected["value"], atol=1e-5)
-    assert np.isclose(entry["std"], expected["std"], atol=1e-5)
-    assert entry["units"] == expected["units"]
+            # Evaluator uses non-mapped SMILES, pseudocode here used mapped
+            # Molecule.from_mapped_smiles(found_smiles)
 
-    assert entry["source"] == expected["source"]
+        assert entry["temperature"] == expected["temperature"]
+        if expected["pressure"] is not None:
+            assert numpy.isclose(entry["pressure"], expected["pressure"], atol=1e-3)
+        else:
+            assert entry["pressure"] is None
 
+        assert numpy.isclose(entry["value"], expected["value"], atol=1e-5)
+        assert numpy.isclose(entry["std"], expected["std"], atol=1e-5)
+        assert Unit(entry["units"]) == Unit(expected["units"])
 
-@pytest.mark.skip(reason="Not implemented yet")
+        assert entry["source"] == expected["source"]
+
+    def test_pandas_roundtrip(self, filename, expected):
+        dataset = ThermoMLDataSet.from_xml(open(get_test_data_path(f"thermoml/{filename}")).read())
+
+        roundtripped = ThermoMLDataSet.from_pandas(dataset.to_pandas())
+
+        assert len(dataset) == len(roundtripped)
+
+
+@pytest.mark.skip(reason="Implement next")
 def test_load_single_osmotic():
     """
     Test loading a single osmotic coefficient data point from a ThermoML XML file.
@@ -111,31 +118,73 @@ def test_load_single_osmotic():
     but is included here to ensure that ions are dealt with correctly.
 
     """
-    dataset = thermoml_dataset_from_xml(get_test_data_path("thermoml/single_osmotic.xml"))
+    dataset = ThermoMLDataSet.from_xml(open(get_test_data_path("thermoml/single_osmotic.xml")).read())
     assert len(dataset) == 1
 
-    entry = dataset[0]
-    assert entry["type"] == "osmotic_coefficient"
+    entry = next(iter(dataset))
 
-    assert "." in entry["smiles_a"]
-    Molecule.from_mapped_smiles(entry["smiles_a"])
-    assert entry["x_a"] == 0.00086
+    assert "." in entry["smiles"]
+    Molecule.from_mapped_smiles(entry["smiles"])
+    assert entry["x"] == 0.00086
 
-    Molecule.from_mapped_smiles(entry["smiles_b"])
-    assert entry["x_b"] == 0.99914
+    Molecule.from_mapped_smiles(entry["smiles"])
+    assert entry["x"] == 0.99914
 
-    assert np.isclose(entry["temperature"], 298.15, atol=1e-3)
+    assert numpy.isclose(entry["temperature"], 298.15, atol=1e-3)
     assert entry["pressure"] is None
-    assert np.isclose(entry["value"], 0.7389, atol=1e-5)
-    assert np.isclose(entry["std"], 0.00655, atol=1e-5)
+    assert numpy.isclose(entry["value"], 0.7389, atol=1e-5)
+    assert numpy.isclose(entry["std"], 0.00655, atol=1e-5)
     assert entry["units"] == "dimensionless"
     assert entry["source"] == "10.1016/j.fluid.2006.09.025"
 
 
-@pytest.mark.skip(reason="Not implemented yet")
 def test_load_from_doi():
     """Test loading a ThermoML dataset from a DOI"""
-    dataset = thermoml_dataset_from_doi("10.1016/j.fluid.2014.12.023")
-    assert len(dataset) == 9
+    dataset = ThermoMLDataSet.from_doi("10.1016/j.fluid.2014.12.023")
+    assert len(dataset) == 186
     for entry in dataset:
         assert entry["source"] == "10.1016/j.fluid.2014.12.023"
+
+
+def test_to_pandas():
+    """A test to ensure that data sets are convertable to pandas objects."""
+
+    thermoml_dataset = ThermoMLDataSet()
+
+    density_entry = {
+        "id": str(uuid.uuid4()).replace("-", ""),
+        "tag": "density",
+        "x": [1.0],
+        "smiles": ["[C:1]([O:5][C:3]([C:2]([O:4][H:13])([H:9])[H:10])([H:11])[H:12])([H:6])([H:7])[H:8]"],
+        "temperature": 293.15,
+        "pressure": 1.0,
+        "value": 0.96488,
+        "std": 0.00005,
+        "units": "g/mL",
+        "source": "",
+    }
+
+    thermoml_dataset.add_properties(density_entry)
+
+    dataframe = thermoml_dataset.to_pandas()
+
+    required_columns = [
+        "Id",
+        "tag",
+        "Temperature (K)",
+        "Pressure (kPa)",
+        "N Components",
+        "Component 1",
+        "Mole Fraction 1",
+        "Value",
+        "Uncertainty",
+        "Source",
+    ]
+
+    assert all(x in dataframe for x in required_columns)
+
+    assert dataframe is not None
+    assert dataframe.shape == (1, 10)
+
+    data_set_without_na = dataframe.dropna(axis=1, how="all")
+    assert data_set_without_na.shape == (1, 9)

dimsim/configs/targets/thermo.py

@@ -0,0 +1,53 @@
+import typing
+
+import datasets
+import pyarrow
+
+from dimsim.molecule import map_smiles
+
+DATA_SCHEMA = pyarrow.schema(
+    [
+        ("id", pyarrow.string()),
+        ("tag", pyarrow.string()),
+        ("smiles", pyarrow.list_(pyarrow.string())),
+        ("x", pyarrow.list_(pyarrow.float64())),
+        ("temperature", pyarrow.float64()),
+        ("pressure", pyarrow.float64()),
+        ("value", pyarrow.float64()),
+        ("std", pyarrow.float64()),
+        ("units", pyarrow.string()),
+        ("source", pyarrow.string()),
+    ]
+)
+
+
+class DataEntry(typing.TypedDict):
+    id: str
+
+    tag: str  # was previously EntryTag
+
+    smiles: list[str]
+
+    x: list[float]
+
+    temperature: float
+
+    pressure: float
+
+    value: float
+
+    std: float | None
+
+    units: str
+
+    source: str
+
+
+def create_dataset(rows: typing.Iterable[DataEntry]) -> datasets.Dataset:
+    for row in rows:
+        row["smiles"] = [map_smiles(value) for value in row["smiles"]]
+
+    # TODO: validate rows
+    table = pyarrow.Table.from_pylist([*rows], schema=DATA_SCHEMA)
+
+    return datasets.Dataset(datasets.table.InMemoryTable(table))