def test_multiple_variables(self):
conditions = [SplitCondition(0, 2, gt), SplitCondition(1, 1, gt)]
X = self.X[:, 0].reshape(3, 2)
combined_condition = CombinedCondition([0, 1], conditions)
self.assertListEqual(list(combined_condition.condition(X)),
[False, True, True])
def test_multiple_conditions(self):
conditions = [SplitCondition(0, 2, gt), SplitCondition(0, 5, le)]
combined_condition = CombinedCondition([0], conditions)
self.assertEqual(combined_condition.variables[0].min_value, 2)
self.assertEqual(combined_condition.variables[0].max_value, 5)
self.assertListEqual(list(combined_condition.condition(self.X)),
[False, False, True, False, True, True])
def test_single_condition_data(self):
data = Data(pd.DataFrame({"a": [1, 2]}).values, np.array([1, 2]))
left_condition, right_condition = SplitCondition(0, 1,
le), SplitCondition(
0, 1, gt)
left_split, right_split = Split(data) + left_condition, Split(
data) + right_condition
self.assertListEqual([1], list(left_split.data.X[:, 0]))
self.assertListEqual([2], list(right_split.data.X[:, 0]))
def test_combined_condition_data(self):
data = make_bartpy_data(
pd.DataFrame({
"a": [1, 2, 3, 4]
}).values, np.array([1, 2, 1, 1]))
first_left_condition, first_right_condition = SplitCondition(
0, 3, le), SplitCondition(0, 3, gt)
second_left_condition, second_right_condition = SplitCondition(
0, 1, le), SplitCondition(0, 1, gt)
split = Split(data)
updated_split = split + first_left_condition + second_right_condition
conditioned_data = updated_split.data
self.assertListEqual([2, 3], list(conditioned_data.X.get_column(0)))
def sample_split_condition(node: LeafNode) -> Optional[Tuple[SplitCondition, SplitCondition]]:
"""
Randomly sample a splitting rule for a particular leaf node
Works based on two random draws
- draw a node to split on based on multinomial distribution
- draw an observation within that variable to split on
Returns None if there isn't a possible non-degenerate split
"""
split_variable = np.random.choice(list(node.split.data.splittable_variables()))
split_value = node.data.random_splittable_value(split_variable)
if split_value is None:
return None
return SplitCondition(split_variable, split_value, le), SplitCondition(split_variable, split_value, gt)
def test_most_recent_split(self):
data = make_bartpy_data(
pd.DataFrame({
"a": [1, 2, 3, 4]
}).values, np.array([1, 2, 1, 1]))
first_left_condition, first_right_condition = SplitCondition(
0, 3, le), SplitCondition(0, 3, gt)
second_left_condition, second_right_condition = SplitCondition(
0, 1, le), SplitCondition(0, 1, gt)
split = Split(data)
updated_split = split + first_left_condition + second_right_condition
self.assertEqual(
(split + first_left_condition).most_recent_split_condition(),
first_left_condition)
self.assertEqual(updated_split.most_recent_split_condition(),
second_right_condition)
def setUp(self):
self.data = Data(
pd.DataFrame({
"a": [1, 2, 3],
"b": [1, 2, 3]
}).values, np.array([1, 2, 3]))
self.a = split_node(LeafNode(Split(
self.data)), (SplitCondition(0, 1, le), SplitCondition(0, 1, gt)))
self.b = self.a.left_child
self.x = self.a.right_child
self.tree = Tree([self.a, self.b, self.x])
self.c = split_node(
self.a._right_child,
(SplitCondition(1, 2, le), SplitCondition(1, 2, gt)))
mutate(self.tree, TreeMutation("grow", self.x, self.c))
self.d = self.c.left_child
self.e = self.c.right_child
def setUp(self):
X = format_covariate_matrix(
pd.DataFrame({
"a": [1, 2, 3],
"b": [1, 2, 3]
}))
self.data = Data(X, np.array([1, 2, 3]).astype(float))
self.a = split_node(LeafNode(Split(
self.data)), (SplitCondition(0, 1, le), SplitCondition(0, 1, gt)))
self.b = self.a.left_child
self.x = self.a.right_child
self.tree = Tree([self.a, self.b, self.x])
self.c = split_node(
self.a._right_child,
(SplitCondition(1, 2, le), SplitCondition(1, 2, gt)))
mutate(self.tree, TreeMutation("grow", self.x, self.c))
self.d = self.c.left_child
self.e = self.c.right_child
def test_single_condition(self):
condition = SplitCondition(0, 3, gt)
combined_condition = CombinedCondition([0], [condition])
self.assertListEqual(list(combined_condition.condition(self.X)),
[False, False, True, True, False, True])