def test_boston_dataset(n_bins):
X, y = load_boston(return_X_y=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)
mapper = _BinMapper(n_bins=n_bins, random_state=42)
X_train_binned = mapper.fit_transform(X_train)
# Init gradients and hessians to that of least squares loss
gradients = -y_train.astype(G_H_DTYPE)
hessians = np.ones(1, dtype=G_H_DTYPE)
min_samples_leaf = 8
max_leaf_nodes = 31
grower = TreeGrower(X_train_binned,
gradients,
hessians,
min_samples_leaf=min_samples_leaf,
max_leaf_nodes=max_leaf_nodes,
n_bins=n_bins,
n_bins_non_missing=mapper.n_bins_non_missing_)
grower.grow()
predictor = grower.make_predictor(bin_thresholds=mapper.bin_thresholds_)
assert r2_score(y_train, predictor.predict(X_train)) > 0.85
assert r2_score(y_test, predictor.predict(X_test)) > 0.70
def test_min_samples_leaf(n_samples, min_samples_leaf, n_bins,
constant_hessian, noise):
rng = np.random.RandomState(seed=0)
# data = linear target, 3 features, 1 irrelevant.
X = rng.normal(size=(n_samples, 3))
y = X[:, 0] - X[:, 1]
if noise:
y_scale = y.std()
y += rng.normal(scale=noise, size=n_samples) * y_scale
mapper = _BinMapper(n_bins=n_bins)
X = mapper.fit_transform(X)
all_gradients = y.astype(G_H_DTYPE)
shape_hessian = 1 if constant_hessian else all_gradients.shape
all_hessians = np.ones(shape=shape_hessian, dtype=G_H_DTYPE)
grower = TreeGrower(X, all_gradients, all_hessians,
n_bins=n_bins, shrinkage=1.,
min_samples_leaf=min_samples_leaf,
max_leaf_nodes=n_samples)
grower.grow()
predictor = grower.make_predictor(
bin_thresholds=mapper.bin_thresholds_)
if n_samples >= min_samples_leaf:
for node in predictor.nodes:
if node['is_leaf']:
assert node['count'] >= min_samples_leaf
else:
assert predictor.nodes.shape[0] == 1
assert predictor.nodes[0]['is_leaf']
assert predictor.nodes[0]['count'] == n_samples
def test_missing_value_predict_only():
# Make sure that missing values are supported at predict time even if they
# were not encountered in the training data: the missing values are
# assigned to whichever child has the most samples.
rng = np.random.RandomState(0)
n_samples = 100
X_binned = rng.randint(0, 256, size=(n_samples, 1), dtype=np.uint8)
X_binned = np.asfortranarray(X_binned)
gradients = rng.normal(size=n_samples).astype(G_H_DTYPE)
hessians = np.ones(shape=1, dtype=G_H_DTYPE)
grower = TreeGrower(X_binned, gradients, hessians, min_samples_leaf=5,
has_missing_values=False)
grower.grow()
predictor = grower.make_predictor()
# go from root to a leaf, always following node with the most samples.
# That's the path nans are supposed to take
node = predictor.nodes[0]
while not node['is_leaf']:
left = predictor.nodes[node['left']]
right = predictor.nodes[node['right']]
node = left if left['count'] > right['count'] else right
prediction_main_path = node['value']
# now build X_test with only nans, and make sure all predictions are equal
# to prediction_main_path
all_nans = np.full(shape=(n_samples, 1), fill_value=np.nan)
assert np.all(predictor.predict(all_nans) == prediction_main_path)
def test_init_parameters_validation():
X_binned, all_gradients, all_hessians = _make_training_data()
with pytest.raises(ValueError,
match="min_gain_to_split=-1 must be positive"):
TreeGrower(X_binned, all_gradients, all_hessians,
min_gain_to_split=-1)
with pytest.raises(ValueError,
match="min_hessian_to_split=-1 must be positive"):
TreeGrower(X_binned, all_gradients, all_hessians,
min_hessian_to_split=-1)
def test_input_validation():
X_binned, all_gradients, all_hessians = _make_training_data()
X_binned_float = X_binned.astype(np.float32)
with pytest.raises(NotImplementedError,
match="X_binned must be of type uint8"):
TreeGrower(X_binned_float, all_gradients, all_hessians)
X_binned_C_array = np.ascontiguousarray(X_binned)
with pytest.raises(
ValueError,
match="X_binned should be passed as Fortran contiguous array"):
TreeGrower(X_binned_C_array, all_gradients, all_hessians)
def test_max_depth(max_depth):
# Make sure max_depth parameter works as expected
rng = np.random.RandomState(seed=0)
n_bins = 256
n_samples = 1000
# data = linear target, 3 features, 1 irrelevant.
X = rng.normal(size=(n_samples, 3))
y = X[:, 0] - X[:, 1]
mapper = _BinMapper(n_bins=n_bins)
X = mapper.fit_transform(X)
all_gradients = y.astype(G_H_DTYPE)
all_hessians = np.ones(shape=1, dtype=G_H_DTYPE)
grower = TreeGrower(X, all_gradients, all_hessians, max_depth=max_depth)
grower.grow()
depth = max(leaf.depth for leaf in grower.finalized_leaves)
assert depth == max_depth
def test_split_on_nan_with_infinite_values():
# Make sure the split on nan situations are respected even when there are
# samples with +inf values (we set the threshold to +inf when we have a
# split on nan so this test makes sure this does not introduce edge-case
# bugs). We need to use the private API so that we can also test
# predict_binned().
X = np.array([0, 1, np.inf, np.nan, np.nan]).reshape(-1, 1)
# the gradient values will force a split on nan situation
gradients = np.array([0, 0, 0, 100, 100], dtype=G_H_DTYPE)
hessians = np.ones(shape=1, dtype=G_H_DTYPE)
bin_mapper = _BinMapper()
X_binned = bin_mapper.fit_transform(X)
n_bins_non_missing = 3
has_missing_values = True
grower = TreeGrower(X_binned, gradients, hessians,
n_bins_non_missing=n_bins_non_missing,
has_missing_values=has_missing_values,
min_samples_leaf=1)
grower.grow()
predictor = grower.make_predictor(
bin_thresholds=bin_mapper.bin_thresholds_
)
# sanity check: this was a split on nan
assert predictor.nodes[0]['threshold'] == np.inf
assert predictor.nodes[0]['bin_threshold'] == n_bins_non_missing - 1
# Make sure in particular that the +inf sample is mapped to the left child
# Note that lightgbm "fails" here and will assign the inf sample to the
# right child, even though it's a "split on nan" situation.
predictions = predictor.predict(X)
predictions_binned = predictor.predict_binned(
X_binned, missing_values_bin_idx=bin_mapper.missing_values_bin_idx_)
assert np.all(predictions == -gradients)
assert np.all(predictions_binned == -gradients)
def test_predictor_from_grower():
# Build a tree on the toy 3-leaf dataset to extract the predictor.
n_bins = 256
X_binned, all_gradients, all_hessians = _make_training_data(
n_bins=n_bins)
grower = TreeGrower(X_binned, all_gradients, all_hessians,
n_bins=n_bins, shrinkage=1.,
max_leaf_nodes=3, min_samples_leaf=5)
grower.grow()
assert grower.n_nodes == 5 # (2 decision nodes + 3 leaves)
# Check that the node structure can be converted into a predictor
# object to perform predictions at scale
predictor = grower.make_predictor()
assert predictor.nodes.shape[0] == 5
assert predictor.nodes['is_leaf'].sum() == 3
# Probe some predictions for each leaf of the tree
# each group of 3 samples corresponds to a condition in _make_training_data
input_data = np.array([
[0, 0],
[42, 99],
[128, 254],
[129, 0],
[129, 85],
[254, 85],
[129, 86],
[129, 254],
[242, 100],
], dtype=np.uint8)
missing_values_bin_idx = n_bins - 1
predictions = predictor.predict_binned(input_data, missing_values_bin_idx)
expected_targets = [1, 1, 1, 1, 1, 1, -1, -1, -1]
assert np.allclose(predictions, expected_targets)
# Check that training set can be recovered exactly:
predictions = predictor.predict_binned(X_binned, missing_values_bin_idx)
assert np.allclose(predictions, -all_gradients)
def test_min_samples_leaf_root(n_samples, min_samples_leaf):
# Make sure root node isn't split if n_samples is not at least twice
# min_samples_leaf
rng = np.random.RandomState(seed=0)
n_bins = 256
# data = linear target, 3 features, 1 irrelevant.
X = rng.normal(size=(n_samples, 3))
y = X[:, 0] - X[:, 1]
mapper = _BinMapper(n_bins=n_bins)
X = mapper.fit_transform(X)
all_gradients = y.astype(G_H_DTYPE)
all_hessians = np.ones(shape=1, dtype=G_H_DTYPE)
grower = TreeGrower(X, all_gradients, all_hessians,
n_bins=n_bins, shrinkage=1.,
min_samples_leaf=min_samples_leaf,
max_leaf_nodes=n_samples)
grower.grow()
if n_samples >= min_samples_leaf * 2:
assert len(grower.finalized_leaves) >= 2
else:
assert len(grower.finalized_leaves) == 1
def test_grow_tree(n_bins, constant_hessian, stopping_param, shrinkage):
X_binned, all_gradients, all_hessians = _make_training_data(
n_bins=n_bins, constant_hessian=constant_hessian)
n_samples = X_binned.shape[0]
if stopping_param == "max_leaf_nodes":
stopping_param = {"max_leaf_nodes": 3}
else:
stopping_param = {"min_gain_to_split": 0.01}
grower = TreeGrower(X_binned, all_gradients, all_hessians,
n_bins=n_bins, shrinkage=shrinkage,
min_samples_leaf=1, **stopping_param)
# The root node is not yet splitted, but the best possible split has
# already been evaluated:
assert grower.root.left_child is None
assert grower.root.right_child is None
root_split = grower.root.split_info
assert root_split.feature_idx == 0
assert root_split.bin_idx == n_bins // 2
assert len(grower.splittable_nodes) == 1
# Calling split next applies the next split and computes the best split
# for each of the two newly introduced children nodes.
left_node, right_node = grower.split_next()
# All training samples have ben splitted in the two nodes, approximately
# 50%/50%
_check_children_consistency(grower.root, left_node, right_node)
assert len(left_node.sample_indices) > 0.4 * n_samples
assert len(left_node.sample_indices) < 0.6 * n_samples
if grower.min_gain_to_split > 0:
# The left node is too pure: there is no gain to split it further.
assert left_node.split_info.gain < grower.min_gain_to_split
assert left_node in grower.finalized_leaves
# The right node can still be splitted further, this time on feature #1
split_info = right_node.split_info
assert split_info.gain > 1.
assert split_info.feature_idx == 1
assert split_info.bin_idx == n_bins // 3
assert right_node.left_child is None
assert right_node.right_child is None
# The right split has not been applied yet. Let's do it now:
assert len(grower.splittable_nodes) == 1
right_left_node, right_right_node = grower.split_next()
_check_children_consistency(right_node, right_left_node, right_right_node)
assert len(right_left_node.sample_indices) > 0.1 * n_samples
assert len(right_left_node.sample_indices) < 0.2 * n_samples
assert len(right_right_node.sample_indices) > 0.2 * n_samples
assert len(right_right_node.sample_indices) < 0.4 * n_samples
# All the leafs are pure, it is not possible to split any further:
assert not grower.splittable_nodes
# Check the values of the leaves:
assert grower.root.left_child.value == approx(shrinkage)
assert grower.root.right_child.left_child.value == approx(shrinkage)
assert grower.root.right_child.right_child.value == approx(-shrinkage,
rel=1e-3)