def test_initialize_error():
with pytest.raises(ValueError):
class BadType:
def __init__(self):
self.bad = True
KDTree.initialize([[0, 0, 0], [1, 1, 1], [0, 2, 0]], accept=BadType)
def test_search_1D(): tree = KDTree([1]) right = KDTree([2]) left = KDTree([0]) tree.right = right tree.left = left assert tree.search([2]) == right assert tree.search([0]) == left assert tree.search([3]) is None assert tree.search([-1]) is None
def test_2NN_2D():
tree = KDTree.initialize([[1, 1], [5, 5], [6, 5]])
nn = tree.nearest_neighbor([4, 5], n=2)
exp = [[[5, 5], 1.0], [[6, 5], 2.0]]
for i in range(len(nn)):
assert np.all(nn[i][0] == np.asarray(exp[i][0]))
assert nn[i][1] == exp[i][1]
def test_init_with_params():
tree = KDTree([0, 0], k=2, axis=1)
assert tree.value == [0, 0]
assert tree.k == 2
assert tree.axis == 1
assert tree.left == None
assert tree.right == None
assert tree.nodes == 1
def test_init():
tree = KDTree(0)
assert tree.value == 0
assert tree.k == 1
assert tree.axis == 0
assert tree.left == None
assert tree.right == None
assert tree.nodes == 1
def test_2NN_accept():
tree = KDTree.initialize(
[KDSubType(1, 1), KDSubType(1, 2),
KDSubType(1, 4)], accept=KDSubType)
nn = tree.nearest_neighbor(KDSubType(1, 2), n=2)
exp = [[KDSubType(1, 2), 0], [KDSubType(1, 1), 1]]
for i in range(len(nn)):
assert np.all(nn[i][0] == exp[i][0])
assert nn[i][1] == exp[i][1]
def test_delete_2D(points_2d_delete, delete_2d, delete_2d_exp, capsys): tree = KDTree.initialize(points_2d_delete) tree = tree.delete(delete_2d) tree.visualize() captured = capsys.readouterr() assert captured.out == delete_2d_exp
def test_2NN():
tree = KDTree.initialize([[1], [5], [6]])
assert np.all(
tree.nearest_neighbor([4], n=2) == np.asarray([[[5], 1], [[6], 2]]))
def test_visualize(points_vis, vis_exp, capsys):
tree = KDTree.initialize(points_vis)
tree.visualize()
captured = capsys.readouterr()
assert captured.out == vis_exp
def test_search_mismatch_accept(): tree = KDTree.initialize([KDSubType(1,0), KDSubType(1,1)], accept=KDSubType) with pytest.raises(ValueError): tree.search(KDSubType(2,0)) with pytest.raises(AttributeError): tree.search(0)
def test_search_mismatch(): tree = KDTree.initialize([[1],[2]]) with pytest.raises(ValueError): tree.search([0,0])
def test_search_accept(): tree = KDTree(KDSubType(1,1), accept=KDSubType) left = KDTree(KDSubType(1,0), accept=KDSubType) tree.left = left assert tree.search(KDSubType(1,0)) == left assert tree.search(KDSubType(1,3)) is None
def test_insert_mismatch():
tree = KDTree.initialize([[1], [2]])
with pytest.raises(ValueError):
tree.insert([0, 0])
def test_dPN_mismatch():
tree = KDTree.initialize([[1], [2]])
with pytest.raises(ValueError):
assert tree.proximal_neighbor([0, 0])
def test_d1PN():
tree = KDTree.initialize([[1], [4]])
assert np.all(tree.proximal_neighbor([3], d=1) == np.asarray([[[4], 1]]))
def test_initialize_accept(points_type, init_type_exp):
tree = KDTree.initialize(points_type, accept=KDSubType)
assert np.all(tree.value == init_type_exp)
assert np.all(tree.right.value == KDSubType(2, [1, 0]))
assert np.all(tree.left.value == KDSubType(2, [-1, 1]))
assert np.all(tree.left.left.value == KDSubType(2, [0, 0]))
def test_initialize_3D(points_3d, init_3d_exp):
tree = KDTree.initialize(points_3d)
assert np.all(tree.value == init_3d_exp)
def test_initialize_2D(points_2d, init_2d_exp, capsys):
tree = KDTree.initialize(points_2d)
tree.visualize()
captured = capsys.readouterr()
assert captured.out == init_2d_exp
def test_initialize_1D(points_1d, init_1d_exp):
tree = KDTree.initialize(points_1d)
assert np.all(tree.value == init_1d_exp)
def test_KNN_mismatch():
tree = KDTree.initialize([[1], [2]])
with pytest.raises(ValueError):
tree.nearest_neighbor([0, 0])
def test_delete_mismatch(): tree = KDTree.initialize([[1],[2]]) with pytest.raises(ValueError): tree.delete([0,0])
def test_d2PN():
tree = KDTree.initialize([[1], [5], [6]])
assert np.all(
tree.proximal_neighbor([4], d=2) == np.asarray([[[5], 1], [[6], 2]]))
def test_search_2D(): tree = KDTree([1,1], k=2) right = KDTree([1,2], k=2, axis=1) tree.right = right assert tree.search([1,2]) == right assert tree.search([1,0]) is None
def test_1NN():
tree = KDTree.initialize([[1], [4]])
assert np.all(tree.nearest_neighbor([3]) == np.asarray([[[4], 1]]))
def test_insert_1D(points_1d_insert, insert_1d, insert_1d_exp, capsys):
tree = KDTree.initialize(points_1d_insert)
tree = tree.insert(insert_1d)
tree.visualize()
captured = capsys.readouterr()
assert captured.out == insert_1d_exp
