def test_mean_centering():
X1 = np.array([[1, 2, 3], [4, 5, 6]])
X1_out = np.array([[-1, -1, -1], [1, 1, 1]])
mc = MeanCenterer()
assert (mc.fit_transform(X1).all() == X1_out.all())
X2 = [1, 2, 3]
X2_out = np.array([-1, 0, 1])
mc = MeanCenterer()
assert (mc.fit_transform(X2).all().all() == X2_out.all())
def test_list_mean_centering():
X2 = [1.0, 2.0, 3.0]
X2_out = np.array([-1.0, 0.0, 1.0])
mc = MeanCenterer()
assert(mc.fit_transform(X2).all().all() == X2_out.all())
def test_array_mean_centering():
X1 = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
X1_out = np.array([[-1.0, -1.0, -1.0], [1.0, 1.0, 1.0]])
mc = MeanCenterer()
assert(mc.fit_transform(X1).all() == X1_out.all())
def test_fitting_error():
X1 = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
mc = MeanCenterer()
mc.transform(X1)
def test_fitting_error():
X1 = np.array([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
mc = MeanCenterer()
with pytest.raises(AttributeError):
mc.transform(X1)