CLN: Fix black/flake8 minor

examples/16_noise_robustness/utils.py

@@ -169,7 +169,6 @@ def make_test_trajectories(
     all_t_test = dict()
 
     for i, equation_name in enumerate(systems_list):
-
         dimension = all_properties[equation_name]["embedding_dimension"]
         all_sols_test[equation_name] = np.zeros((n, n_trajectories, dimension))
         all_t_test[equation_name] = np.zeros((n, n_trajectories))

examples/17_parameterized_pattern_formation/utils.py

@@ -38,7 +38,7 @@ def get_lorenz_trajectories(sigmas, rhos, betas, dt):
             x0_train,
             args=(sigmas[i], betas[i], rhos[i]),
             t_eval=t_train,
-            **integrator_keywords
+            **integrator_keywords,
         ).y.T
         x_trains = x_trains + [x_train]
         t_trains = t_trains + [t_train]
@@ -966,7 +966,7 @@ def oregonator(
                 args=(b,),
                 y0=ut,
                 first_step=dt,
-                **integrator_keywords
+                **integrator_keywords,
             )
             if not usol.success:
                 print(usol.message)
@@ -1164,7 +1164,7 @@ def oregonator_homogeneous(
             args=(b,),
             y0=ut,
             first_step=dt,
-            **integrator_keywords
+            **integrator_keywords,
         )
         dt = np.diff(usol.t)[-1]
         ut = usol.y[:, -1]
@@ -1267,7 +1267,7 @@ def cgle_homogeneous(t, u):
             (t[i], t[i + 1]),
             y0=ut,
             first_step=dt,
-            **integrator_keywords
+            **integrator_keywords,
         )
         #         print(usol.message,end='\r')
         dt = np.diff(usol.t)[-1]

pysindy/differentiation/finite_difference.py

@@ -66,7 +66,6 @@ def __init__(
         drop_endpoints=False,
         periodic=False,
     ):
-
         if order <= 0 or not isinstance(order, int):
             raise ValueError("order must be a positive int")
         if d <= 0:

pysindy/feature_library/polynomial_library.py

@@ -1,6 +1,7 @@
 from itertools import chain
 from math import comb
 from typing import Iterator
+from typing import Tuple
 
 import numpy as np
 from numpy.typing import NDArray
@@ -80,7 +81,7 @@ def _combinations(
         include_interaction: bool,
         interaction_only: bool,
         include_bias: bool,
-    ) -> Iterator[tuple[int, ...]]:
+    ) -> Iterator[Tuple[int, ...]]:
         """
         Create selection tuples of input indexes for each polynomail term
 

pysindy/feature_library/sindy_pi_library.py

@@ -393,7 +393,6 @@ def transform(self, x_full):
                         f_dot.__code__.co_argcount,
                         self.interaction_only,
                     ):
-
                         for i, f in enumerate(self.x_functions):
                             for f_combs in self._combinations(
                                 n_input_features,

pysindy/feature_library/weak_pde_library.py

@@ -457,7 +457,6 @@ def _set_up_weights(self):
         deriv = np.zeros(self.grid_ndim)
         deriv[-1] = 1
         for k in range(self.K):
-
             ret = np.ones(shapes_k[k])
             for i in range(self.grid_ndim):
                 s = [0] * (self.grid_ndim + 1)
@@ -476,7 +475,6 @@ def _set_up_weights(self):
         # Product weights over the axes for pure derivative terms, shaped as inds_k
         self.fullweights0 = []
         for k in range(self.K):
-
             ret = np.ones(shapes_k[k])
             for i in range(self.grid_ndim):
                 s = [0] * (self.grid_ndim + 1)
@@ -922,7 +920,6 @@ def transform(self, x_full):
                         for deriv in derivs[
                             np.where(np.all(derivs <= derivs_mixed, axis=1))[0]
                         ]:
-
                             # indices for product rule terms
                             j0 = np.where(np.all(derivs == deriv, axis=1))[0][0]
                             j1 = np.where(

pysindy/optimizers/sbr.py

@@ -99,7 +99,6 @@ def __init__(
         unbias: bool = False,
         **kwargs,
     ):
-
         if unbias:
             raise ValueError("SBR is incompatible with unbiasing. Set unbias=False")
 

pysindy/optimizers/trapping_sr3.py

@@ -539,7 +539,6 @@ def _reduce(self, x, y):
         # Begin optimization loop
         self.objective_history_ = []
         for k in range(self.max_iter):
-
             # update P tensor from the newest trap center
             mPQ = np.tensordot(trap_ctr, self.PQ_, axes=([0], [0]))
             p = self.PL_ - mPQ

pysindy/utils/axes.py

@@ -430,7 +430,6 @@ def decorator(func):
     return decorator
 
 
-
 @_implements(np.ravel)
 def ravel(a, order="C"):
     out = np.ravel(np.asarray(a), order=order)
@@ -455,7 +454,6 @@ def concatenate(arrays, axis=0, out=None, dtype=None, casting="same_kind"):
     ax_list = [obj.axes for obj in arrays if isinstance(obj, AxesArray)]
     for ax1, ax2 in zip(ax_list[:-1], ax_list[1:]):
         if ax1 != ax2:
-
             raise ValueError("Concatenating >1 AxesArray with incompatible axes")
     result = np.concatenate(parents, axis, out=out, dtype=dtype, casting=casting)
     if isinstance(out, AxesArray):

pysindy/utils/odes.py

@@ -334,6 +334,7 @@ def burgers_galerkin(sigma=0.1, nu=0.025, U=1.0):
 # Below this line are models only suitable for SINDy-PI, since they are implicit #
 # and therefore require a library Theta(X, Xdot) rather than just Theta(X) #
 
+
 # Michaelis–Menten model for enzyme kinetics
 def enzyme(t, x, jx=0.6, Vmax=1.5, Km=0.3):
     return jx - Vmax * x / (Km + x)

test/conftest.py

@@ -60,7 +60,6 @@ def data_1d_bad_shape():
 
 @pytest.fixture(scope="session")
 def data_lorenz():
-
     t = np.linspace(0, 1, 12)
     x0 = [8, 27, -7]
     x = solve_ivp(lorenz, (t[0], t[-1]), x0, t_eval=t).y.T
@@ -70,7 +69,6 @@ def data_lorenz():
 
 @pytest.fixture
 def data_multiple_trajectories():
-
     n_points = [100, 200, 500]
     initial_conditions = [
         [8, 27, -7],
@@ -120,7 +118,6 @@ def diffuse(t, u, dx, nx):
 
 @pytest.fixture(scope="session")
 def data_discrete_time():
-
     n_steps = 100
     mu = 3.6
     x = np.zeros((n_steps))
@@ -133,7 +130,6 @@ def data_discrete_time():
 
 @pytest.fixture(scope="session")
 def data_discrete_time_multiple_trajectories():
-
     n_steps = 100
     mus = [1, 2.3, 3.6]
     x = [np.zeros((n_steps)) for mu in mus]
@@ -421,7 +417,6 @@ def u_fun(t):
 
 @pytest.fixture(scope="session")
 def data_discrete_time_c():
-
     n_steps = 100
     mu = 3.6
 
@@ -437,7 +432,6 @@ def data_discrete_time_c():
 
 @pytest.fixture(scope="session")
 def data_discrete_time_c_multivariable():
-
     n_steps = 100
     mu = 3.6
 
@@ -454,7 +448,6 @@ def data_discrete_time_c_multivariable():
 
 @pytest.fixture(scope="session")
 def data_discrete_time_multiple_trajectories_c():
-
     n_steps = 100
     mus = [1, 2.3, 3.6]
     u = [0.001 * np.random.randn(n_steps) for mu in mus]

test/utils/test_axes.py

@@ -29,13 +29,6 @@ def test_bad_concat():
         np.concatenate((arr, arr2), axis=0)
 
 
-def test_concat_out():
-    arr = AxesArray(np.arange(3).reshape(1, 3), {"ax_a": 0, "ax_b": 1})
-    arr_out = np.empty((2, 3)).view(AxesArray)
-    result = np.concatenate((arr, arr), axis=0, out=arr_out)
-    assert_equal(result, arr_out)
-
-
 def test_reduce_mean_noinf_recursion():
     arr = AxesArray(np.array([[1]]), {"ax_a": [0, 1]})
     np.mean(arr, axis=0)