test.yml remove fail-fast: true, fix some doc strings missing "Return…

…s:", change parse_vasp_dir error type FileNotFoundError->NotADirectoryError

.github/workflows/test.yml

@@ -10,7 +10,6 @@ on:
 jobs:
   tests:
     strategy:
-      fail-fast: true
       matrix:
         os: [ubuntu-latest, macos-14, windows-latest]
         version:

.pre-commit-config.yaml

@@ -4,7 +4,7 @@ default_install_hook_types: [pre-commit, commit-msg]
 
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.6.2
+    rev: v0.6.4
     hooks:
       - id: ruff
         args: [--fix]
@@ -48,7 +48,7 @@ repos:
           - svelte
 
   - repo: https://github.com/pre-commit/mirrors-eslint
-    rev: v9.9.0
+    rev: v9.10.0
     hooks:
       - id: eslint
         types: [file]

chgnet/trainer/trainer.py

@@ -664,8 +664,18 @@ def load(cls, path: str) -> Self:
         return trainer
 
     @staticmethod
-    def move_to(obj, device) -> Tensor | list[Tensor]:
-        """Move object to device."""
+    def move_to(
+        obj: Tensor | list[Tensor], device: torch.device
+    ) -> Tensor | list[Tensor]:
+        """Move object to device.
+
+        Args:
+            obj (Tensor | list[Tensor]): object(s) to move to device
+            device (torch.device): device to move object to
+
+        Returns:
+            Tensor | list[Tensor]: moved object(s)
+        """
         if torch.is_tensor(obj):
             return obj.to(device)
         if isinstance(obj, list):

chgnet/utils/common_utils.py

@@ -39,6 +39,9 @@ def cuda_devices_sorted_by_free_mem() -> list[int]:
     """List available CUDA devices sorted by increasing available memory.
 
     To get the device with the most free memory, use the last list item.
+
+    Returns:
+        list[int]: CUDA device numbers sorted by increasing free memory.
     """
     if not torch.cuda.is_available():
         return []
@@ -94,10 +97,10 @@ def mae(prediction: Tensor, target: Tensor) -> Tensor:
 
 
 def read_json(filepath: str) -> dict:
-    """Read the json file.
+    """Read the JSON file.
 
     Args:
-        filepath (str): file name of json to read.
+        filepath (str): file name of JSON to read.
 
     Returns:
         dict: data stored in filepath
@@ -107,27 +110,27 @@ def read_json(filepath: str) -> dict:
 
 
 def write_json(dct: dict, filepath: str) -> dict:
-    """Write the json file.
+    """Write the JSON file.
 
     Args:
         dct (dict): dictionary to write
-        filepath (str): file name of json to write.
-
-    Returns:
-        written dictionary
+        filepath (str): file name of JSON to write.
     """
 
     def handler(obj: object) -> int | object:
         """Convert numpy int64 to int.
 
         Fixes TypeError: Object of type int64 is not JSON serializable
         reported in https://github.com/CederGroupHub/chgnet/issues/168.
+
+        Returns:
+            int | object: object for serialization
         """
         if isinstance(obj, np.integer):
             return int(obj)
         return obj
 
-    with open(filepath, "w") as file:
+    with open(filepath, mode="w") as file:
         json.dump(dct, file, default=handler)
 
 

chgnet/utils/vasp_utils.py

@@ -30,9 +30,16 @@ def parse_vasp_dir(
             Exception to VASP calculation that did not achieve electronic convergence.
             Default = True
         save_path (str): path to save the parsed VASP labels
+
+    Raises:
+        NotADirectoryError: if the base_dir is not a directory
+
+    Returns:
+        dict: a dictionary of lists with keys for structure, uncorrected_total_energy,
+            energy_per_atom, force, magmom, stress.
     """
     if os.path.isdir(base_dir) is False:
-        raise FileNotFoundError(f"{base_dir=} is not a directory")
+        raise NotADirectoryError(f"{base_dir=} is not a directory")
 
     oszicar_path = zpath(f"{base_dir}/OSZICAR")
     vasprun_path = zpath(f"{base_dir}/vasprun.xml")
@@ -170,6 +177,9 @@ def solve_charge_by_mag(
                     (3.5, 4.2): 3,
                     (4.2, 5): 2
                 ))
+
+    Returns:
+        Structure: pymatgen Structure with oxidation states assigned based on magmoms.
     """
     out_structure = structure.copy()
     out_structure.remove_oxidation_states()

examples/make_graphs.py

@@ -58,8 +58,22 @@ def make_graphs(
     make_partition(labels, graph_dir, train_ratio, val_ratio)
 
 
-def make_one_graph(mp_id: str, graph_id: str, data, graph_dir) -> dict | bool:
-    """Convert a structure to a CrystalGraph and save it."""
+def make_one_graph(
+    mp_id: str, graph_id: str, data: StructureJsonData, graph_dir: str
+) -> dict | bool:
+    """Convert a structure to a CrystalGraph and save it.
+
+    Args:
+        mp_id (str): The material id.
+        graph_id (str): The graph id.
+        data (StructureJsonData): The dataset. Warning: Dicts are popped from the data,
+            i.e. modifying the data in place.
+        graph_dir (str): The directory to save the graphs.
+
+    Returns:
+        dict | bool: The label dictionary if the graph is saved successfully, False
+            otherwise.
+    """
     dct = data.data[mp_id].pop(graph_id)
     struct = Structure.from_dict(dct.pop("structure"))
     try:

pyproject.toml

@@ -102,6 +102,7 @@ docstring-code-format = true
     "ANN201",
     "D100",
     "D103",
+    "DOC201",  # doc string missing Return section
     "FBT001",
     "FBT002",
     "INP001",

site/make_docs.py

@@ -49,5 +49,5 @@
     markdown = markdown.replace(
         "\n**Global Variables**\n---------------\n- **TYPE_CHECKING**\n\n", ""
     )
-    with open(path, "w") as file:
+    with open(path, mode="w") as file:
         file.write(markdown)

tests/test_dataset.py

@@ -20,23 +20,17 @@
 def structure_data() -> StructureData:
     """Create a graph with 3 nodes and 3 directed edges."""
     random.seed(42)
-    structures, energies, forces, stresses, magmoms, structure_ids = (
-        [],
-        [],
-        [],
-        [],
-        [],
-        [],
-    )
+    structures, energies, forces = [], [], []
+    stresses, magmoms, structure_ids = [], [], []
+
     for index in range(100):
-        struct = NaCl.copy()
-        struct.perturb(0.1)
-        structures.append(struct)
-        energies.append(np.random.random(1))
-        forces.append(np.random.random([2, 3]))
-        stresses.append(np.random.random([3, 3]))
-        magmoms.append(np.random.random([2, 1]))
-        structure_ids.append(index)
+        structures += [NaCl.copy().perturb(0.1)]
+        energies += [np.random.random(1)]
+        forces += [np.random.random([2, 3])]
+        stresses += [np.random.random([3, 3])]
+        magmoms += [np.random.random([2, 1])]
+        structure_ids += [index]
+
     return StructureData(
         structures=structures,
         energies=energies,