def test_invert(self):
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
b = a * a.invert()
np.testing.assert_array_almost_equal(b.numpy(), np.eye(3))
a = tf2d.from_elements([0] * 8 + [1])
self.assertRaises(ValueError, tf2d.invert, a)
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
b = a * ~a
np.testing.assert_array_almost_equal(b.numpy(), np.eye(3))
a = tf2d(np.arange(9).reshape(3, 3))
self.assertRaises(ValueError, a.invert)
def test_translate(self):
tx, ty = 1.5, 2.5
b = tf2d().translate(tx, ty)
A = np.eye(3)
A[0, 2] = tx
A[1, 2] = ty
np.testing.assert_array_almost_equal(b.numpy(), A)
def test_rotate(self):
theta = 35 * np.pi / 180
b = tf2d().rotate(theta, degrees=False)
M = b.numpy()
self.assertEqual(M[0, 0], M[1, 1])
self.assertEqual(M[1, 0], -M[0, 1])
self.assertEqual(np.arccos(M[0, 0]), theta)
self.assertEqual(np.arcsin(M[1, 0]), theta)
def test_is_degenerate(self):
a = tf2d(np.arange(9).reshape(3, 3))
self.assertEqual(True, a.is_degenerate())
def test_det(self):
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
self.assertEqual(a.det(), 6)
def test_copy(self):
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
b = a.copy()
np.testing.assert_array_almost_equal(a.numpy(), b.numpy())
def test_reset(self):
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
a.reset()
np.testing.assert_array_almost_equal(a.numpy(), np.eye(3))
def test_init(self):
A = [1, 0.1, 3, 0, 1, 4, 0.1, 0.2, 1.05]
A = np.array(A).reshape(3, 3)
A[2, 2] = 1
M = tf2d(A).numpy()
np.testing.assert_array_almost_equal(M, A)
def test_apply(self):
p = np.array([[0, 0], [0, 1], [1, 0]])
a = tf2d().rotate(180).translate(1, 1).scale(2, 3)
q = a * p
r = np.array([[2., 3.], [2., 0.], [0., 3.]])
np.testing.assert_array_almost_equal(r, q)
def test_mul(self):
a = tf2d().rotate(30).translate(2, 2)
b = tf2d().shear(10, 20).scale(1, 2)
c = a * b
np.testing.assert_array_almost_equal(c.numpy(), a.numpy() @ b.numpy())
def test_shear(self):
theta_x, theta_y = 35 * np.pi / 180, 65 * np.pi / 180
b = tf2d().shear(theta_x, theta_y, degrees=False)
self.assertEqual(np.arctan(b.numpy()[0, 1]), theta_x)
self.assertEqual(np.arctan(b.numpy()[1, 0]), theta_y)
def test_scale(self):
s_x, s_y = 1.5, 2.5
b = tf2d().scale(s_x, s_y)
self.assertEqual(b.numpy()[0, 0], s_x)
self.assertEqual(b.numpy()[1, 1], s_y)