format improvements

cat_scaling/data/eads.py

@@ -83,7 +83,9 @@ def data(
             try:
                 data = data.astype(float)
             except ValueError as e:
-                raise ValueError(f"Please double-check input data: {e}.")
+                raise ValueError(
+                    "Please double-check input data for Eads."
+                ) from e
 
         else:
             raise TypeError("Expect data as pd.DataFrame type.")
@@ -239,14 +241,13 @@ def remove_sample(
 
     def sort_data(
         self,
-        targets: list[str] = ["column", "row"],
+        targets: set[str] | None = None,
     ) -> None:
         """Sort columns/rows of data."""
+        if targets is None:
+            targets = {"column", "row"}
 
-        if not set(targets) <= {
-            "column",
-            "row",
-        }:
+        elif targets > {"column", "row"}:
             raise ValueError(
                 "Invalid target values. Should be 'column', 'row', or both."
             )

cat_scaling/data/reaction.py

@@ -70,20 +70,14 @@ def __str__(self) -> str:
         # Add reactants
         reactants = []
         for spec, num in self.reactants.items():
-            if spec.adsorbed:
-                name = f"*{spec.name}"
-            else:
-                name = spec.name
+            name = f"*{spec.name}" if spec.adsorbed else spec.name
 
             reactants.append(f"{num}{name}")
 
         # Add products
         products = []
         for spec, num in self.products.items():
-            if spec.adsorbed:
-                name = f"*{spec.name}"
-            else:
-                name = spec.name
+            name = f"*{spec.name}" if spec.adsorbed else spec.name
 
             products.append(f"{num}{name}")
 
@@ -150,10 +144,9 @@ def _sepa_stoi_number(name: str) -> tuple[float, str]:
         name = name.strip()
 
         # Use re to separate leading digits and name
-        match = re.match(r"^(\d+(\.\d+)?)(.*)$", name)
-        if match:
-            stoi_number_str = match.group(1)
-            species_name = match.group(3)
+        if match := re.match(r"^(\d+(\.\d+)?)(.*)$", name):
+            stoi_number_str = match[1]
+            species_name = match[3]
 
         else:
             stoi_number_str = ""

cat_scaling/data/species.py

@@ -136,7 +136,7 @@ def from_str(cls, string: str) -> Self:
         string = string.strip()
 
         # Check if adsorbed
-        adsorbed: bool = True if string.startswith("*") else False
+        adsorbed: bool = bool(string.startswith("*"))
 
         # Get energy and correction
         e_start = string.find("(")

cat_scaling/plotters/volcanos.py

@@ -33,7 +33,7 @@ def VolcanoPlotter2D(
     y = np.linspace(*y_range)
 
     # Generate limiting potential mesh
-    # TODO: indexes should be used for rate-determining step (RDS)
+    # TODO: indexes should be used for rate-determining step (RDS) plot
     limit_potentials, _rds = relation.eval_limit_potential_2D(x, y)
 
     # Generate limiting potential volcano plot

cat_scaling/relation/builder.py

@@ -88,10 +88,7 @@ def _build_composite_descriptor(
 
         # Fetch child descriptors
         child_descriptors = np.array(
-            [
-                self.data.get_adsorbate(species)
-                for species in spec_ratios.keys()
-            ]
+            [self.data.get_adsorbate(species) for species in spec_ratios]
         )
 
         # Construct composite descriptor (from child descriptors)
@@ -272,7 +269,7 @@ def build_adaptive(
             warnings.warn("Small step length may slow down searching.")
 
         # Convert step_length to percentage
-        step_length = step_length / 100
+        step_length /= 100
 
         # Get descriptors as a list of names
         _descriptors = descriptors.descriptors
@@ -304,8 +301,8 @@ def build_adaptive(
 
                 scores[ratio] = metrics
 
-            # Rerun linear regression with the optimal ratio
-            opt_ratio = max(scores, key=lambda k: scores[k])
+            # Find and rerun linear regression with the optimal ratio
+            opt_ratio = max(scores, key=scores.get)  # type: ignore[arg-type]
 
             opt_ratios = {
                 _descriptors[0]: opt_ratio,

cat_scaling/relation/descriptors.py

@@ -7,6 +7,17 @@
 
 
 class Descriptors:
+    """Helper class to record descriptors.
+
+    Attributes:
+        groups (dict): A dictionary representing groups of adsorbates
+            and their respective descriptors.
+            Keys are descriptors and values are lists of group members.
+            If a group has no members, its value is None.
+        method (str, optional): The method used for building Relation.
+            Should be either "traditional" or "adaptive".
+    """
+
     def __init__(self, groups: dict, method: Optional[str] = None) -> None:
         self.groups = groups
         self.method = method
@@ -81,7 +92,10 @@ def method(self) -> Optional[str]:
 
     @method.setter
     def method(self, method: Optional[str]):
-        if method is not None and method.lower() not in {"traditional", "adaptive"}:
+        if method is not None and method.lower() not in {
+            "traditional",
+            "adaptive",
+        }:
             raise ValueError("Invalid method.")
 
         self._method = method.lower() if method is not None else None

cat_scaling/relation/relation.py

@@ -279,7 +279,7 @@ def coefficients(self, coefficients: list[np.ndarray]):
             raise TypeError("All coefficients should be numpy arrays.")
 
         # Check if all arrays have the same length
-        if len(set(arr.shape[0] for arr in coefficients)) > 1:
+        if len({arr.shape[0] for arr in coefficients}) > 1:
             raise ValueError(
                 "All coefficient arrays should have the same length."
             )
@@ -331,7 +331,6 @@ def _eval_deltaE_2D(
             # Extract coefficients
             coef_x, coef_y, intercept = coef
 
-            # TODO: more efficient method/algorithm?
             return xx * coef_x + yy * coef_y + intercept
 
         def _eval_limit_potential_2D(

tests/data/test_eads.py

@@ -76,7 +76,7 @@ def test_remove_sample(self, setup_class):
         assert "Cu@g-C3N4" not in self.eads.samples
 
     def test_sort_df(self, setup_class):
-        self.eads.sort_data(targets=["column", "row"])
+        self.eads.sort_data(targets={"column", "row"})
 
         assert self.eads.adsorbates == [
             "*CO",

tests/relation/test_descriptors.py

@@ -35,5 +35,7 @@ def test_member_overlap(self):
             "*OH": ["*OCH3", "*O", "*CH2O"],
         }
 
-        with pytest.warns(UserWarning, match="Descriptor group members overlap."):
+        with pytest.warns(
+            UserWarning, match="Descriptor group members overlap."
+        ):
             Descriptors(groups, method="traditional")