def test_get_sklearn_nodes_from_weighted_X(pseudo_tree):
X = np.array([[1, 0, 1, 0], [0, 1, 1, 0], [1, 1, 1, 0], [0, 0, 0, 0],
[1, 0, 1, 1], [0, 1, 1, 1], [1, 1, 1, 1], [0, 0, 0, 1]])
nodes = get_sklearn_nodes_from(pseudo_tree, X=X, weighted=True)
expected_nodes = [
Node(index=0,
left=1,
right=4,
feature=0,
value=[.7, .3],
impurity=.42,
n_node_samples=10.4),
Node(index=1,
left=2,
right=3,
feature=1,
value=[.7, .3],
impurity=.42,
n_node_samples=5.2),
Leaf(index=2, value=[1., 0.], impurity=0., n_node_samples=2.),
Leaf(index=3, value=[0., 1.], impurity=0., n_node_samples=4.),
Node(index=4,
left=5,
right=6,
feature=2,
value=[.7, .3],
impurity=.42,
n_node_samples=5.2),
Leaf(index=5, value=[1., 0.], impurity=0., n_node_samples=0.),
Leaf(index=6, value=[0., 1.], impurity=0., n_node_samples=8.)
]
assert nodes == expected_nodes
def test_get_sklearn_nodes_from_non_weighted(pseudo_tree):
nodes = get_sklearn_nodes_from(pseudo_tree, weighted=False)
expected_nodes = [
Node(index=0,
left=1,
right=4,
feature=0,
value=[.7, .3],
impurity=.42,
n_node_samples=100),
Node(index=1,
left=2,
right=3,
feature=1,
value=[.7, .3],
impurity=.42,
n_node_samples=50),
Leaf(index=2, value=[1., 0.], impurity=0., n_node_samples=35),
Leaf(index=3, value=[0., 1.], impurity=0., n_node_samples=15),
Node(index=4,
left=5,
right=6,
feature=2,
value=[.7, .3],
impurity=.42,
n_node_samples=50),
Leaf(index=5, value=[1., 0.], impurity=0., n_node_samples=35),
Leaf(index=6, value=[0., 1.], impurity=0., n_node_samples=15)
]
assert nodes == expected_nodes
def test__compute_impurity_importance_from():
node = Node(index=1,
left=2,
right=3,
feature=0,
value=[.7, .3],
impurity=.42,
n_node_samples=100)
left = Leaf(index=2, value=[1., 0.], impurity=0., n_node_samples=70)
right = Leaf(index=3, value=[0., 1.], impurity=0., n_node_samples=30)
impurity_importance = MeanDecreaseImpurity._compute_impurity_importance_from(
node, left, right)
expected_impurity_importance = 42.
assert math.isclose(impurity_importance, expected_impurity_importance)
def test_leaf_initialization_attributes(leaf_initializers):
leaf = Leaf(**leaf_initializers)
for attribute, value in leaf_initializers.items():
assert getattr(leaf, attribute) == value
assert math.isclose(
leaf.weighted_impurity,
leaf_initializers['n_node_samples'] * leaf_initializers['impurity'])
def test_leaf_initialization(leaf_initializers):
leaf = Leaf(**leaf_initializers)
assert hasattr(leaf, 'index')
assert hasattr(leaf, 'value')
assert hasattr(leaf, 'impurity')
assert hasattr(leaf, 'n_node_samples')
assert hasattr(leaf, 'weighted_impurity')
def test__get_node_from(node_initializers):
node = _get_node_from(**node_initializers)
expected_node = Node(**node_initializers)
assert node == expected_node
assert not (node is expected_node)
leaf_initializers = copy.deepcopy(node_initializers)
leaf_initializers['left'], leaf_initializers['right'] = -1, -1
leaf = _get_node_from(**leaf_initializers)
del leaf_initializers['left']
del leaf_initializers['right']
del leaf_initializers['feature']
expected_leaf = Leaf(**leaf_initializers)
assert leaf == expected_leaf
assert not (leaf is expected_leaf)
def test_leaf_inequality(leaf_initializers, leaf_initializers_for_inequality):
leaf1, leaf2 = Leaf(**leaf_initializers), Leaf(
**leaf_initializers_for_inequality)
assert leaf1 != leaf2
assert not (leaf1 is leaf2)
def test_leaf_equality(leaf_initializers):
leaf1, leaf2 = Leaf(**leaf_initializers), Leaf(**leaf_initializers)
assert leaf1 == leaf2
assert not (leaf1 is leaf2)