def test_mtry(self, lung_X, lung_y, mtry):
forest = GRFForestSurvival(mtry=mtry)
forest.fit(lung_X, lung_y)
if mtry is not None:
assert forest.mtry_ == mtry
else:
assert forest.mtry_ == 3
def test_fit(self, lung_X, lung_y):
forest = GRFForestSurvival()
with pytest.raises(NotFittedError):
check_is_fitted(forest)
forest.fit(lung_X, lung_y)
check_is_fitted(forest)
assert hasattr(forest, "grf_forest_")
assert hasattr(forest, "mtry_")
assert forest.criterion == "logrank"
def test_with_X_nan(self, lung_X, lung_y):
lung_X_nan = lung_X.copy()
index = np.random.choice(lung_X_nan.size, 100, replace=False)
lung_X_nan = lung_X_nan.to_numpy()
shape = lung_X_nan.shape
lung_X_nan = lung_X_nan.ravel()
lung_X_nan[index] = np.nan
lung_X_nan = lung_X_nan.reshape(shape)
assert np.sum(np.isnan(lung_X_nan)) == 100
forest = GRFForestSurvival()
forest.fit(lung_X_nan, lung_y)
pred = forest.predict(lung_X_nan)
assert len(pred) == lung_X_nan.shape[0]
def test_serialize(self, lung_X, lung_y):
forest = GRFForestSurvival()
# not fitted
tf = tempfile.TemporaryFile()
pickle.dump(forest, tf)
tf.seek(0)
forest = pickle.load(tf)
forest.fit(lung_X, lung_y)
# fitted
tf = tempfile.TemporaryFile()
pickle.dump(forest, tf)
tf.seek(0)
new_forest = pickle.load(tf)
pred = new_forest.predict(lung_X)
assert len(pred) == lung_X.shape[0]
def test_estimators_(self, lung_X, lung_y):
forest = GRFForestSurvival(n_estimators=10)
with pytest.raises(AttributeError):
_ = forest.estimators_
forest.fit(lung_X, lung_y)
with pytest.raises(ValueError):
_ = forest.estimators_
forest = GRFForestSurvival(n_estimators=10, enable_tree_details=True)
forest.fit(lung_X, lung_y)
estimators = forest.estimators_
assert len(estimators) == 10
assert isinstance(estimators[0], GRFTreeSurvival)
check_is_fitted(estimators[0])
def test_alpha(self, lung_X, lung_y, alpha):
forest = GRFForestSurvival(alpha=alpha)
if alpha <= 0 or alpha >= 0.25:
with pytest.raises(ValueError):
forest.fit(lung_X, lung_y)
else:
forest.fit(lung_X, lung_y)
def test_sample_fraction(self, lung_X, lung_y, sample_fraction):
forest = GRFForestSurvival(sample_fraction=sample_fraction)
if sample_fraction <= 0 or sample_fraction > 1:
with pytest.raises(ValueError):
forest.fit(lung_X, lung_y)
else:
forest.fit(lung_X, lung_y)
def test_honesty_fraction(self, lung_X, lung_y, honesty_fraction):
forest = GRFForestSurvival(
honesty=True, honesty_fraction=honesty_fraction, honesty_prune_leaves=True
)
if honesty_fraction <= 0 or honesty_fraction >= 1:
with pytest.raises(RuntimeError):
forest.fit(lung_X, lung_y)
else:
forest.fit(lung_X, lung_y)
def test_get_kernel_weights(self, lung_X, lung_y):
X_train, X_test, y_train, y_test = train_test_split(
lung_X, lung_y, test_size=0.33, random_state=42
)
forest = GRFForestSurvival()
forest.fit(X_train, y_train)
weights = forest.get_kernel_weights(X_test)
assert weights.shape[0] == X_test.shape[0]
assert weights.shape[1] == X_train.shape[0]
oob_weights = forest.get_kernel_weights(X_train, True)
assert oob_weights.shape[0] == X_train.shape[0]
assert oob_weights.shape[1] == X_train.shape[0]
def test_equalize_cluster_weights(
self, lung_X, lung_y, lung_cluster, equalize_cluster_weights
):
forest = GRFForestSurvival(equalize_cluster_weights=equalize_cluster_weights)
forest.fit(lung_X, lung_y, cluster=lung_cluster)
if equalize_cluster_weights:
assert forest.samples_per_cluster_ == 20
else:
assert forest.samples_per_cluster_ == lung_y.shape[0] - 20
if equalize_cluster_weights:
with pytest.raises(ValueError):
forest.fit(
lung_X,
lung_y,
cluster=lung_cluster,
sample_weight=np.ones(lung_y.shape),
)
forest.fit(lung_X, lung_y, cluster=None)
assert forest.samples_per_cluster_ == 0
def test_get_feature_importances(self, lung_X, lung_y):
forest = GRFForestSurvival()
forest.fit(lung_X, lung_y)
fi = forest.get_feature_importances()
assert len(fi) == lung_X.shape[1]
def test_clone(self, lung_X, lung_y):
forest = GRFForestSurvival()
forest.fit(lung_X, lung_y)
clone(forest)
def test_honesty_prune_leaves(self, lung_X, lung_y, honesty_prune_leaves):
forest = GRFForestSurvival(
honesty=True, honesty_prune_leaves=honesty_prune_leaves
)
forest.fit(lung_X, lung_y)
def test_get_tags(self):
forest = GRFForestSurvival()
tags = forest._get_tags()
assert tags["requires_y"]
def test_predict(self, lung_X, lung_y):
forest = GRFForestSurvival()
forest.fit(lung_X, lung_y)
pred = forest.predict(lung_X)
assert len(pred) == lung_X.shape[0]
def test_init(self):
_ = GRFForestSurvival()
def test_get_split_frequencies(self, lung_X, lung_y):
forest = GRFForestSurvival()
forest.fit(lung_X, lung_y)
sf = forest.get_split_frequencies()
assert sf.shape[1] == lung_X.shape[1]
def test_get_estimator(self, lung_X, lung_y):
forest = GRFForestSurvival(n_estimators=10)
with pytest.raises(NotFittedError):
_ = forest.get_estimator(idx=0)
forest.fit(lung_X, lung_y)
with pytest.raises(ValueError):
_ = forest.get_estimator(idx=0)
forest = GRFForestSurvival(n_estimators=10, enable_tree_details=True)
forest.fit(lung_X, lung_y)
estimator = forest.get_estimator(0)
check_is_fitted(estimator)
assert isinstance(estimator, GRFTreeSurvival)
with pytest.raises(IndexError):
_ = forest.get_estimator(idx=20)
def test_from_forest(self, lung_X, lung_y):
forest = GRFForestSurvival()
forest.fit(lung_X, lung_y)
tree = GRFTreeSurvival.from_forest(forest=forest, idx=0)
tree.predict(lung_X)
def test_honesty(self, lung_X, lung_y, honesty):
forest = GRFForestSurvival(honesty=honesty)
forest.fit(lung_X, lung_y)