TST: Break apart integration test for nan derivatives.

Also, make test/differentiation/test_differentiation_methods.py just
test/test_differentiation.py.  Unneccessary nesting.

Previously, test_nan_derivatives asked whether the model fit by SINDy when
FiniteDifference gave NaN endpointswas close to the model fit without nans.
This is an integration test that is really asking whether FiniteDifference is
calculating derivatives in a consistent with drop_endpoints=True and False.

Also, a bit faster but this test was fast to begin with.

test/differentiation/test_differentiation_methods.py → test/test_differentiation.py

@@ -45,9 +45,12 @@ def test_centered_difference_variable_timestep_length():
     assert len(centered_difference(x, t) == len(x))
 
 
-# Fixtures: data sets to be re-used in multiple tests
-# data_derivative_1d and data_derivative_2d are defined
-# in ../conftest.py
+def test_nan_derivatives(data_lorenz):
+    x, t = data_lorenz
+    x_dot = FiniteDifference(drop_endpoints=False)(x, t)
+    x_dot_nans = FiniteDifference(drop_endpoints=True)(x, t)
+    np.testing.assert_allclose(x_dot_nans[1:-1], x_dot[1:-1])
+    assert np.isnan(x_dot_nans[:1]).all() and np.isnan(x_dot_nans[-1:]).all()
 
 
 def test_forward_difference_1d(data_derivative_1d):

test/test_pysindy.py

@@ -23,7 +23,6 @@
 from sklearn.utils.validation import check_is_fitted
 
 from pysindy import SINDy
-from pysindy.differentiation import FiniteDifference
 from pysindy.differentiation import SINDyDerivative
 from pysindy.differentiation import SmoothedFiniteDifference
 from pysindy.feature_library import FourierLibrary
@@ -81,20 +80,6 @@ def test_improper_shape_input(data_1d):
     check_is_fitted(model)
 
 
-def test_nan_derivatives(data_lorenz):
-    x, t = data_lorenz
-    model0 = SINDy(differentiation_method=FiniteDifference(drop_endpoints=False))
-    model0.fit(x, t)
-
-    model = SINDy(differentiation_method=FiniteDifference(drop_endpoints=True))
-    model.fit(x, t)
-    check_is_fitted(model)
-
-    result = model.score(x, t=t)
-    expected = model0.score(x, t=t)
-    np.testing.assert_allclose(result, expected, rtol=1e-4)
-
-
 @pytest.mark.parametrize(
     "data",
     [