def test_mtry(self, boston_X, boston_y, mtry):
tree = GRFTreeQuantileRegressor(mtry=mtry)
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
if mtry is not None:
assert tree.mtry_ == mtry
else:
assert tree.mtry_ == 6
def test_with_X_nan(self, boston_X, boston_y):
boston_X_nan = boston_X.copy()
index = np.random.choice(boston_X_nan.size, 100, replace=False)
boston_X_nan.ravel()[index] = np.nan
assert np.sum(np.isnan(boston_X_nan)) == 100
tree = GRFTreeQuantileRegressor(quantiles=[0.5])
tree.fit(boston_X_nan, boston_y)
pred = tree.predict(boston_X_nan)
assert len(pred) == boston_X_nan.shape[0]
def test_serialize(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2, 0.5, 0.8]
# not fitted
tf = tempfile.TemporaryFile()
pickle.dump(tree, tf)
tf.seek(0)
tree = pickle.load(tf)
tree.fit(boston_X, boston_y)
# fitted
tf = tempfile.TemporaryFile()
pickle.dump(tree, tf)
tf.seek(0)
new_tree = pickle.load(tf)
pred = new_tree.predict(boston_X)
assert len(pred) == boston_X.shape[0]
def test_equalize_cluster_weights(self, boston_X, boston_y, boston_cluster,
equalize_cluster_weights):
tree = GRFTreeQuantileRegressor(
equalize_cluster_weights=equalize_cluster_weights)
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y, cluster=boston_cluster)
if equalize_cluster_weights:
assert tree.samples_per_cluster_ == 20
else:
assert tree.samples_per_cluster_ == boston_y.shape[0] - 20
tree.fit(boston_X, boston_y, cluster=boston_cluster)
tree.fit(boston_X, boston_y, cluster=None)
assert tree.samples_per_cluster_ == 0
def test_alpha(self, boston_X, boston_y, alpha):
tree = GRFTreeQuantileRegressor(alpha=alpha)
tree.quantiles = [0.2, 0.5, 0.8]
if alpha <= 0 or alpha >= 0.25:
with pytest.raises(ValueError):
tree.fit(boston_X, boston_y)
else:
tree.fit(boston_X, boston_y)
def test_sample_fraction(self, boston_X, boston_y, sample_fraction):
tree = GRFTreeQuantileRegressor(sample_fraction=sample_fraction)
tree.quantiles = [0.2, 0.5, 0.8]
if sample_fraction <= 0 or sample_fraction > 1:
with pytest.raises(ValueError):
tree.fit(boston_X, boston_y)
else:
tree.fit(boston_X, boston_y)
def test_tree_interface(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2]
tree.fit(boston_X, boston_y)
# access attributes the way we would expect to in sklearn
tree_ = tree.tree_
children_left = tree_.children_left
children_right = tree_.children_right
children_default = tree_.children_default
feature = tree_.feature
threshold = tree_.threshold
max_depth = tree_.max_depth
n_node_samples = tree_.n_node_samples
weighted_n_node_samples = tree_.weighted_n_node_samples
node_count = tree_.node_count
capacity = tree_.capacity
n_outputs = tree_.n_outputs
n_classes = tree_.n_classes
value = tree_.value
def test_honesty_fraction(self, boston_X, boston_y, honesty_fraction):
tree = GRFTreeQuantileRegressor(honesty=True,
honesty_fraction=honesty_fraction,
honesty_prune_leaves=True)
tree.quantiles = [0.2, 0.5, 0.8]
if honesty_fraction <= 0 or honesty_fraction >= 1:
with pytest.raises(RuntimeError):
tree.fit(boston_X, boston_y)
else:
tree.fit(boston_X, boston_y)
def test_get_n_leaves(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2]
tree.fit(boston_X, boston_y)
leaves = tree.get_n_leaves()
assert isinstance(leaves, int)
assert np.all(leaves > 0)
def test_get_depth(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2]
tree.fit(boston_X, boston_y)
depth = tree.get_depth()
assert isinstance(depth, int)
assert depth > 0
def test_decision_path(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2]
tree.fit(boston_X, boston_y)
paths = tree.decision_path(boston_X)
assert isinstance(paths, csr_matrix)
assert paths.shape[0] == len(boston_X)
def test_apply(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2]
tree.fit(boston_X, boston_y)
leaves = tree.apply(boston_X)
assert isinstance(leaves, np.ndarray)
assert np.all(leaves > 0)
assert len(leaves) == len(boston_X)
def test_fit(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
with pytest.raises(NotFittedError):
check_is_fitted(tree)
with pytest.raises(ValueError):
tree.fit(boston_X, boston_y)
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
check_is_fitted(tree)
assert hasattr(tree, "grf_forest_")
assert hasattr(tree, "mtry_")
assert tree.grf_forest_["num_trees"] == 1
assert tree.criterion == "gini"
def test_honesty(self, boston_X, boston_y, honesty):
tree = GRFTreeQuantileRegressor(honesty=honesty)
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
def test_honesty_prune_leaves(self, boston_X, boston_y,
honesty_prune_leaves):
tree = GRFTreeQuantileRegressor(
honesty=True, honesty_prune_leaves=honesty_prune_leaves)
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
def test_check_estimator(self):
check_estimator(GRFTreeQuantileRegressor(quantiles=[0.2]))
def test_clone(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
clone(tree)
def test_from_forest(self, boston_X, boston_y):
forest = GRFForestQuantileRegressor(quantiles=[0.2])
forest.fit(boston_X, boston_y)
tree = GRFTreeQuantileRegressor.from_forest(forest=forest, idx=0)
tree.predict(boston_X)
def test_init(self):
_ = GRFTreeQuantileRegressor()
def test_predict(self, boston_X, boston_y):
tree = GRFTreeQuantileRegressor()
tree.quantiles = [0.2, 0.5, 0.8]
tree.fit(boston_X, boston_y)
pred = tree.predict(boston_X)
assert len(pred) == boston_X.shape[0]