[MegaLinter] Apply linters automatic fixes

src/pyrecest/_backend/_shared_numpy/linalg.py

@@ -110,12 +110,10 @@ def fractional_matrix_power(A, t):
 
 def polar(a, side="right"):
     """Polar decomposition of a square or rectangular matrix."""
-    signature = (
-        "(m,n)->(m,n),(m,m)" if side == "left" else "(m,n)->(m,n),(n,n)"
+    signature = "(m,n)->(m,n),(m,m)" if side == "left" else "(m,n)->(m,n),(n,n)"
+    return _np.vectorize(_scipy.linalg.polar, signature=signature, excluded=["side"])(
+        a, side=side
     )
-    return _np.vectorize(
-        _scipy.linalg.polar, signature=signature, excluded=["side"]
-    )(a, side=side)
 
 
 def solve(a, b):

src/pyrecest/_backend/pytorch/linalg.py

@@ -255,12 +255,8 @@ def fractional_matrix_power(A, t):
 def polar(a, side="right"):
     """Polar decomposition of a square or rectangular matrix."""
     a = _as_linalg_tensor(a)
-    signature = (
-        "(m,n)->(m,n),(m,m)" if side == "left" else "(m,n)->(m,n),(n,n)"
-    )
-    func = _np.vectorize(
-        _scipy.linalg.polar, signature=signature, excluded=["side"]
-    )
+    signature = "(m,n)->(m,n),(m,m)" if side == "left" else "(m,n)->(m,n),(n,n)"
+    func = _np.vectorize(_scipy.linalg.polar, signature=signature, excluded=["side"])
     unitary, hermitian = func(_as_numpy_no_grad(a), side=side)
 
     return _torch_as_like(unitary, a), _torch_as_like(hermitian, a)

src/pyrecest/distributions/abstract_grid_distribution.py

@@ -121,7 +121,9 @@ def multiply(self, other):
         if self.grid_type != other.grid_type:
             raise ValueError("Grid types must match before multiplication.")
         if (self.grid is None) != (other.grid is None):
-            raise ValueError("Both grid distributions must either store grids or omit them.")
+            raise ValueError(
+                "Both grid distributions must either store grids or omit them."
+            )
         if self.grid is not None and not allclose(self.grid, other.grid):
             raise ValueError("Grid coordinates must match before multiplication.")
         gd = copy.deepcopy(self)

src/pyrecest/distributions/hypersphere_subset/complex_angular_central_gaussian_distribution.py

@@ -101,7 +101,9 @@ def pdf(self, za):
         """
         za = array(za)
         if za.ndim == 0 or za.shape[-1] != self.dim:
-            raise ValueError(f"za must have trailing dimension {self.dim}, got {za.shape}.")
+            raise ValueError(
+                f"za must have trailing dimension {self.dim}, got {za.shape}."
+            )
         single = za.ndim == 1
         if single:
             za = za.reshape(1, -1)

src/pyrecest/distributions/hypersphere_subset/complex_watson_distribution.py

@@ -148,7 +148,9 @@ def pdf(self, za):
         """
         za = asarray(za, dtype=complex128)
         if za.ndim == 0 or za.shape[-1] != self.dim:
-            raise ValueError(f"za must have trailing dimension {self.dim}, got {za.shape}.")
+            raise ValueError(
+                f"za must have trailing dimension {self.dim}, got {za.shape}."
+            )
         inner = za @ conj(self.mu)  # complex inner product, shape (N,) or scalar
         return real(exp(self.log_norm_const + self.kappa * abs(inner) ** 2))
 

tests/distributions/test_abstract_grid_distribution.py

@@ -3,7 +3,6 @@
 from pyrecest.backend import array
 from pyrecest.distributions.abstract_grid_distribution import AbstractGridDistribution
 
-
 _DEFAULT_GRID = object()
 
 
@@ -62,26 +61,20 @@ def test_multiply_rejects_incompatible_grids(self):
 
         with self.assertRaisesRegex(ValueError, "enforce_pdf_nonnegative"):
             dist.multiply(
-                DummyGridDistribution(
-                    array([1.0, 1.0]), enforce_pdf_nonnegative=False
-                )
+                DummyGridDistribution(array([1.0, 1.0]), enforce_pdf_nonnegative=False)
             )
 
         with self.assertRaisesRegex(ValueError, "Grid value shapes"):
             dist.multiply(DummyGridDistribution(array([[1.0, 1.0]])))
 
         with self.assertRaisesRegex(ValueError, "Grid types"):
             dist.multiply(
-                DummyGridDistribution(
-                    array([1.0, 1.0]), grid_type="cartesian_prod"
-                )
+                DummyGridDistribution(array([1.0, 1.0]), grid_type="cartesian_prod")
             )
 
         with self.assertRaisesRegex(ValueError, "Grid coordinates"):
             dist.multiply(
-                DummyGridDistribution(
-                    array([1.0, 1.0]), grid=array([[0.0], [2.0]])
-                )
+                DummyGridDistribution(array([1.0, 1.0]), grid=array([[0.0], [2.0]]))
             )
 
     def test_multiply_combines_compatible_values(self):

tests/distributions/test_abstract_sphere_subset_distribution.py

@@ -144,9 +144,7 @@ def test_elevation_helpers_reject_matrix_inputs(self):
         z = array([[0.0, 0.0]])
 
         with self.assertRaisesRegex(ValueError, "1-dimensional"):
-            AbstractSphereSubsetDistribution._sph_to_cart_elevation(
-                azimuth, elevation
-            )
+            AbstractSphereSubsetDistribution._sph_to_cart_elevation(azimuth, elevation)
         with self.assertRaisesRegex(ValueError, "1-dimensional"):
             AbstractSphereSubsetDistribution._cart_to_sph_elevation(x, y, z)
 

tests/distributions/test_complex_angular_central_gaussian_distribution.py

@@ -54,7 +54,9 @@ def test_constructor_rejects_invalid_shape_or_nonfinite_matrix(self):
         ]
 
         for C_bad, message in invalid_matrices:
-            with self.subTest(message=message), self.assertRaisesRegex(ValueError, message):
+            with self.subTest(message=message), self.assertRaisesRegex(
+                ValueError, message
+            ):
                 ComplexAngularCentralGaussianDistribution(C_bad)
 
     def test_constructor_rejects_non_positive_definite_matrix(self):
@@ -142,7 +144,9 @@ def test_pdf_batch_vs_single(self):
     def test_pdf_accepts_list_and_rejects_wrong_dimension(self):
         z = [[1.0 + 0j, 0.0 + 0j]]
 
-        npt.assert_allclose(self.dist_identity_2d.pdf(z), self.dist_identity_2d.pdf(array(z)))
+        npt.assert_allclose(
+            self.dist_identity_2d.pdf(z), self.dist_identity_2d.pdf(array(z))
+        )
 
         for invalid_z in (1.0 + 0j, [1.0 + 0j], [[1.0 + 0j, 0.0 + 0j, 0.0 + 0j]]):
             with self.subTest(invalid_z=invalid_z):

tests/distributions/test_complex_watson_distribution.py

@@ -276,9 +276,7 @@ def test_fit_with_weights(self):
         self.assertIsInstance(cw_fit, ComplexWatsonDistribution)
 
     def test_estimate_parameters_rejects_invalid_inputs(self):
-        valid_Z = array(
-            [[1.0 + 0j, 0.0 + 0j], [0.0 + 0j, 1.0 + 0j]], dtype=complex128
-        )
+        valid_Z = array([[1.0 + 0j, 0.0 + 0j], [0.0 + 0j, 1.0 + 0j]], dtype=complex128)
         invalid_cases = [
             ([1.0 + 0j, 0.0 + 0j], None, "two-dimensional"),
             (array([[float("nan") + 0j, 0.0 + 0j]], dtype=complex128), None, "finite"),

tests/distributions/test_custom_hyperrectangular_distribution.py

@@ -143,7 +143,9 @@ def test_normalize_rejects_non_numpy_backend(self):
             with self.assertRaisesRegex(NotImplementedError, "numpy backend"):
                 dist.normalize()
         finally:
-            backend.__backend_name__ = original_backend_name  # pylint: disable=no-member
+            backend.__backend_name__ = (
+                original_backend_name  # pylint: disable=no-member
+            )
 
 
 if __name__ == "__main__":

tests/distributions/test_hypercylindrical_dirac_distribution.py

@@ -168,9 +168,7 @@ def test_plot(self, name, dist, dim):
         )
 
     def test_plot_rejects_unsupported_dimension(self):
-        dist = HypercylindricalDiracDistribution(
-            1, array([[0.1, 1.0, 2.0, 3.0]])
-        )
+        dist = HypercylindricalDiracDistribution(1, array([[0.1, 1.0, 2.0, 3.0]]))
 
         with self.assertRaisesRegex(ValueError, "Plotting"):
             dist.plot()

tests/distributions/test_hyperhemispherical_watson_distribution.py

@@ -1,7 +1,6 @@
 import unittest
 
 import numpy.testing as npt
-
 from pyrecest.backend import array
 from pyrecest.distributions.hypersphere_subset.hyperhemispherical_watson_distribution import (
     HyperhemisphericalWatsonDistribution,

tests/evaluation/test_pareto_objective_validation.py

@@ -1,13 +1,14 @@
 import pandas as pd
 import pytest
-
 from pyrecest.evaluation.pareto import is_pareto_front, pareto_front_indices
 
 
 def test_pareto_front_indices_reject_missing_objective_column():
     table = pd.DataFrame({"cost": [1.0, 2.0]})
 
-    with pytest.raises(ValueError, match="Pareto objective columns are missing: quality"):
+    with pytest.raises(
+        ValueError, match="Pareto objective columns are missing: quality"
+    ):
         pareto_front_indices(
             table,
             ["cost", "quality"],
@@ -18,7 +19,9 @@ def test_pareto_front_indices_reject_missing_objective_column():
 def test_is_pareto_front_rejects_missing_objective_column():
     table = pd.DataFrame({"cost": [1.0, 2.0]})
 
-    with pytest.raises(ValueError, match="Pareto objective columns are missing: quality"):
+    with pytest.raises(
+        ValueError, match="Pareto objective columns are missing: quality"
+    ):
         is_pareto_front(
             table,
             ["cost", "quality"],

tests/filters/test_measurement_reliability_zero_noise.py

@@ -1,12 +1,15 @@
 import numpy as np
-
-from pyrecest.filters.measurement_reliability import normalize_measurement_noise_covariances
+from pyrecest.filters.measurement_reliability import (
+    normalize_measurement_noise_covariances,
+)
 
 
 def _as_covariance_matrix(value, measurement_dim, name):
     covariance = np.asarray(value, dtype=float)
     if covariance.shape != (measurement_dim, measurement_dim):
-        raise ValueError(f"{name} must have shape ({measurement_dim}, {measurement_dim})")
+        raise ValueError(
+            f"{name} must have shape ({measurement_dim}, {measurement_dim})"
+        )
     return covariance