Skip to content

TST use global_random_seed in sklearn/ensemble/tests/test_iforest.py #22901

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 14 commits into from
May 18, 2022
Merged

TST use global_random_seed in sklearn/ensemble/tests/test_iforest.py #22901

Changes from all commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
29d1282
add global-random-seed in tests
MaxwellLZH Mar 19, 2022
d81556c
remove global_random_seed for test_iforest_with_uniform_data
MaxwellLZH Mar 28, 2022
8c776c2
update test_iforest_performance
MaxwellLZH Mar 28, 2022
85de1cd
Merge remote-tracking branch 'upstream/main' into pr/MaxwellLZH/22901
jeremiedbb May 4, 2022
768d4f2
[all random seeds]
jeremiedbb May 4, 2022
0b1c259
use global_random_seed in iforest construction
MaxwellLZH May 4, 2022
3a956c1
does not use global random state in test functions
MaxwellLZH May 6, 2022
29842dd
clean up code in train_test_split
MaxwellLZH May 6, 2022
feee7e7
clean up code in train_test_split
MaxwellLZH May 6, 2022
e21f7f7
clean up train_test_split & fix failing tests
MaxwellLZH May 6, 2022
66fb5ef
remove global random vareiable
MaxwellLZH May 8, 2022
39c932c
Use the global_random_seed fixture also in test_iforest_chunks_works*
ogrisel May 17, 2022
0fa49bc
[all random seeds]
ogrisel May 17, 2022
bca0de4
[all random seeds]
ogrisel May 17, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
81 changes: 36 additions & 45 deletions sklearn/ensemble/tests/test_iforest.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,24 +27,13 @@
from scipy.sparse import csc_matrix, csr_matrix
from unittest.mock import Mock, patch

rng = check_random_state(0)

# load the iris dataset
# and randomly permute it
# load iris & diabetes dataset
iris = load_iris()
perm = rng.permutation(iris.target.size)
iris.data = iris.data[perm]
iris.target = iris.target[perm]

# also load the diabetes dataset
# and randomly permute it
diabetes = load_diabetes()
perm = rng.permutation(diabetes.target.size)
diabetes.data = diabetes.data[perm]
diabetes.target = diabetes.target[perm]


def test_iforest():
def test_iforest(global_random_seed):
"""Check Isolation Forest for various parameter settings."""
X_train = np.array([[0, 1], [1, 2]])
X_test = np.array([[2, 1], [1, 1]])
Expand All @@ -55,15 +44,15 @@ def test_iforest():

with ignore_warnings():
for params in grid:
IsolationForest(random_state=rng, **params).fit(X_train).predict(X_test)
IsolationForest(random_state=global_random_seed, **params).fit(
X_train
).predict(X_test)


def test_iforest_sparse():
def test_iforest_sparse(global_random_seed):
"""Check IForest for various parameter settings on sparse input."""
rng = check_random_state(0)
X_train, X_test, y_train, y_test = train_test_split(
diabetes.data[:50], diabetes.target[:50], random_state=rng
)
rng = check_random_state(global_random_seed)
X_train, X_test = train_test_split(diabetes.data[:50], random_state=rng)
grid = ParameterGrid({"max_samples": [0.5, 1.0], "bootstrap": [True, False]})

for sparse_format in [csc_matrix, csr_matrix]:
Expand All @@ -73,13 +62,13 @@ def test_iforest_sparse():
for params in grid:
# Trained on sparse format
sparse_classifier = IsolationForest(
n_estimators=10, random_state=1, **params
n_estimators=10, random_state=global_random_seed, **params
).fit(X_train_sparse)
sparse_results = sparse_classifier.predict(X_test_sparse)

# Trained on dense format
dense_classifier = IsolationForest(
n_estimators=10, random_state=1, **params
n_estimators=10, random_state=global_random_seed, **params
).fit(X_train)
dense_results = dense_classifier.predict(X_test)

Expand Down Expand Up @@ -143,41 +132,39 @@ def test_max_samples_attribute():
assert clf.max_samples_ == 0.4 * X.shape[0]


def test_iforest_parallel_regression():
def test_iforest_parallel_regression(global_random_seed):
"""Check parallel regression."""
rng = check_random_state(0)
rng = check_random_state(global_random_seed)

X_train, X_test, y_train, y_test = train_test_split(
diabetes.data, diabetes.target, random_state=rng
)
X_train, X_test = train_test_split(diabetes.data, random_state=rng)

ensemble = IsolationForest(n_jobs=3, random_state=0).fit(X_train)
ensemble = IsolationForest(n_jobs=3, random_state=global_random_seed).fit(X_train)

ensemble.set_params(n_jobs=1)
y1 = ensemble.predict(X_test)
ensemble.set_params(n_jobs=2)
y2 = ensemble.predict(X_test)
assert_array_almost_equal(y1, y2)

ensemble = IsolationForest(n_jobs=1, random_state=0).fit(X_train)
ensemble = IsolationForest(n_jobs=1, random_state=global_random_seed).fit(X_train)

y3 = ensemble.predict(X_test)
assert_array_almost_equal(y1, y3)


def test_iforest_performance():
def test_iforest_performance(global_random_seed):
"""Test Isolation Forest performs well"""

# Generate train/test data
rng = check_random_state(2)
X = 0.3 * rng.randn(120, 2)
X_train = np.r_[X + 2, X - 2]
X_train = X[:100]
rng = check_random_state(global_random_seed)
X = 0.3 * rng.randn(600, 2)
X = rng.permutation(np.vstack((X + 2, X - 2)))
X_train = X[:1000]

# Generate some abnormal novel observations
X_outliers = rng.uniform(low=-4, high=4, size=(20, 2))
X_test = np.r_[X[100:], X_outliers]
y_test = np.array([0] * 20 + [1] * 20)
X_outliers = rng.uniform(low=-1, high=1, size=(200, 2))
X_test = np.vstack((X[1000:], X_outliers))
y_test = np.array([0] * 200 + [1] * 200)

# fit the model
clf = IsolationForest(max_samples=100, random_state=rng).fit(X_train)
Expand All @@ -190,12 +177,12 @@ def test_iforest_performance():


@pytest.mark.parametrize("contamination", [0.25, "auto"])
def test_iforest_works(contamination):
def test_iforest_works(contamination, global_random_seed):
# toy sample (the last two samples are outliers)
X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1], [6, 3], [-4, 7]]
X = [[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1], [7, 4], [-5, 9]]

# Test IsolationForest
clf = IsolationForest(random_state=rng, contamination=contamination)
clf = IsolationForest(random_state=global_random_seed, contamination=contamination)
clf.fit(X)
decision_func = -clf.decision_function(X)
pred = clf.predict(X)
Expand Down Expand Up @@ -281,25 +268,29 @@ def test_iforest_warm_start():


# mock get_chunk_n_rows to actually test more than one chunk (here one
# chunk = 3 rows:
# chunk has 3 rows):
@patch(
"sklearn.ensemble._iforest.get_chunk_n_rows",
side_effect=Mock(**{"return_value": 3}),
)
@pytest.mark.parametrize("contamination, n_predict_calls", [(0.25, 3), ("auto", 2)])
def test_iforest_chunks_works1(mocked_get_chunk, contamination, n_predict_calls):
test_iforest_works(contamination)
def test_iforest_chunks_works1(
mocked_get_chunk, contamination, n_predict_calls, global_random_seed
):
test_iforest_works(contamination, global_random_seed)
assert mocked_get_chunk.call_count == n_predict_calls


# idem with chunk_size = 5 rows
# idem with chunk_size = 10 rows
@patch(
"sklearn.ensemble._iforest.get_chunk_n_rows",
side_effect=Mock(**{"return_value": 10}),
)
@pytest.mark.parametrize("contamination, n_predict_calls", [(0.25, 3), ("auto", 2)])
def test_iforest_chunks_works2(mocked_get_chunk, contamination, n_predict_calls):
test_iforest_works(contamination)
def test_iforest_chunks_works2(
mocked_get_chunk, contamination, n_predict_calls, global_random_seed
):
test_iforest_works(contamination, global_random_seed)
assert mocked_get_chunk.call_count == n_predict_calls


Expand Down