[MegaLinter] Apply linters automatic fixes

src/pyrecest/distributions/hypertorus/abstract_hypertoroidal_distribution.py

@@ -180,7 +180,9 @@ def hellinger_dist_fun(*args):
             x = array(args)
             return (sqrt(self.pdf(x)) - sqrt(other.pdf(x))) ** 2
 
-        squared_dist = 0.5 * self.integrate_fun_over_domain(hellinger_dist_fun, self.dim)
+        squared_dist = 0.5 * self.integrate_fun_over_domain(
+            hellinger_dist_fun, self.dim
+        )
         return sqrt(squared_dist)
 
     def total_variation_distance_numerical(self, other):

src/pyrecest/filters/_ukf.py

@@ -147,9 +147,7 @@ def update(self, z, R=None, hx=None, **hx_args):
         sigmas_h = None
         dim_z = None
         for i in range(sigmas_f.shape[0]):
-            sigma_h = reshape(
-                asarray(hx(sigmas_f[i], **hx_args), dtype=float64), (-1,)
-            )
+            sigma_h = reshape(asarray(hx(sigmas_f[i], **hx_args), dtype=float64), (-1,))
             if sigmas_h is None:
                 dim_z = sigma_h.shape[0]
                 sigmas_h = empty((sigmas_f.shape[0], dim_z))
@@ -179,9 +177,7 @@ def update(self, z, R=None, hx=None, **hx_args):
                 if using_default_R
                 else "measurement noise covariance R"
             )
-            raise ValueError(
-                f"{source} has shape {R.shape}, expected {(dim_z, dim_z)}"
-            )
+            raise ValueError(f"{source} has shape {R.shape}, expected {(dim_z, dim_z)}")
 
         z_pred = einsum("i,ij->j", Wm, sigmas_h)
         Pz, Pxz = self._innovation_matrices(sigmas_f, sigmas_h, z_pred, R, Wc)
@@ -201,4 +197,4 @@ def __deepcopy__(self, memo):
         memo[id(self)] = result
         for k, v in self.__dict__.items():
             setattr(result, k, deepcopy(v, memo))
-        return result
+        return result

src/pyrecest/filters/abstract_particle_filter.py

@@ -25,7 +25,9 @@ def _call_vectorized_sample_next(sample_next, particles, n_particles):
             return sample_next(particles, n_particles)
         return sample_next(particles, n=n_particles)
 
-    if any(param.kind == inspect.Parameter.VAR_KEYWORD for param in parameters.values()):
+    if any(
+        param.kind == inspect.Parameter.VAR_KEYWORD for param in parameters.values()
+    ):
         return sample_next(particles, n=n_particles)
 
     return sample_next(particles)

src/pyrecest/filters/unscented_kalman_filter.py

@@ -34,7 +34,11 @@ def __init__(
             raise ValueError(
                 "initial_state must be a GaussianDistribution or a tuple of (mean, covariance)"
             )
-        dim_z = reshape(asarray(hx(initial_mean), dtype=float64), (-1,)).shape[0] if dim_z is None else dim_z
+        dim_z = (
+            reshape(asarray(hx(initial_mean), dtype=float64), (-1,)).shape[0]
+            if dim_z is None
+            else dim_z
+        )
 
         if points is None:
             # Standard settings for Gaussian approximations

tests/filters/test_unscented_kalman_filter.py

@@ -76,9 +76,7 @@ def test_update_linear_rectangular_measurement(self):
         kf.update_linear(array([1.0]), array([[1.0, 0.0]]), array([[0.5]]))
 
         npt.assert_allclose(kf.get_point_estimate(), array([2.0 / 3.0, 1.0]))
-        npt.assert_allclose(
-            kf.filter_state.covariance(), diag(array([1.0 / 3.0, 2.0]))
-        )
+        npt.assert_allclose(kf.filter_state.covariance(), diag(array([1.0 / 3.0, 2.0])))
 
     @unittest.skipIf(
         pyrecest.backend.__backend_name__ in ("pytorch", "jax"),