def test_OutSamplerTransformer_regressor(multi_output):
np.random.seed(123)
X = np.random.randn(100, 10)
if multi_output:
y = np.random.randn(100, 2)
else:
y = np.random.randn(100)
model = OutSamplerTransformer(RandomForestRegressor(n_estimators=10,
random_state=123),
cv=10)
model.fit(X, y)
y1 = model.model.predict(X)
y2 = model.transform(X)
assert not is_classifier(model)
assert not is_regressor(model)
if multi_output:
assert np.abs(y1[:, 0] - y2[:, 0]).max() <= 10**(-10)
assert np.abs(y1[:, 1] - y2[:, 1]).max() <= 10**(-10)
assert y2.shape == (100, 2)
assert model.get_feature_names() == [
"output%d__RandomForestRegressor__target" % d
for d in range(y.shape[1])
]
else:
assert np.abs(y1 - y2[:, 0]).max() <= 10**(-10)
assert y2.shape == (100, 1)
assert model.get_feature_names() == ["RandomForestRegressor__target"]
def test_OutSamplerTransformer_classifier():
np.random.seed(123)
X = np.random.randn(100, 10)
y = 1 * (np.random.randn(100) > 0)
model = OutSamplerTransformer(
RandomForestClassifier(n_estimators=10, random_state=123))
model.fit(X, y)
p1 = model.model.predict_proba(X)
p2 = model.transform(X)
assert not is_classifier(model)
assert not is_regressor(model)
assert np.abs(p1[:, 1] - p2[:, 0]).max() <= 10**(-10)
assert p2.shape == (100, 1)
assert model.get_feature_names() == ["RandomForestClassifier__1"]
y = np.array(["a", "b", "c"])[np.random.randint(0, 3, 100)]
model = OutSamplerTransformer(
RandomForestClassifier(n_estimators=10, random_state=123))
model.fit(X, y)
p1 = model.model.predict_proba(X)
p2 = model.transform(X)
assert p1.shape == (100, 3)
assert p2.shape == (100, 3)
assert np.abs(p1 - p2).max() <= 10**(-10)
assert model.get_feature_names() == [
"RandomForestClassifier__a",
"RandomForestClassifier__b",
"RandomForestClassifier__c",
]
def test_OutSamplerTransformer_regressor():
np.random.seed(123)
X = np.random.randn(100, 10)
y = np.random.randn(100)
model = OutSamplerTransformer(RandomForestRegressor(n_estimators=10,
random_state=123),
cv=10)
model.fit(X, y)
y1 = model.model.predict(X)
y2 = model.transform(X)
assert not is_classifier(model)
assert not is_regressor(model)
assert np.abs(y1 - y2[:, 0]).max() <= 10**(-10)
assert y2.shape == (100, 1)
assert model.get_feature_names() == ["RandomForestRegressor__target"]
def test_OutSamplerTransformer_classifier(multi_output):
np.random.seed(123)
X = np.random.randn(100, 10)
if multi_output:
y = 1 * (np.random.randn(100, 2) > 0)
else:
y = 1 * (np.random.randn(100) > 0)
model = OutSamplerTransformer(
RandomForestClassifier(n_estimators=10, random_state=123))
model.fit(X, y)
p1 = model.model.predict_proba(X)
p2 = model.transform(X)
assert not is_classifier(model)
assert not is_regressor(model)
if multi_output:
assert np.abs(p1[0][:, 1] - p2[:, 0]).max() <= 10**(-10)
assert np.abs(p1[1][:, 1] - p2[:, 1]).max() <= 10**(-10)
else:
assert np.abs(p1[:, 1] - p2[:, 0]).max() <= 10**(-10)
assert p2.shape == (100, 1 + 1 * multi_output)
if multi_output:
assert model.get_feature_names() == [
"output%d__RandomForestClassifier__1" % d
for d in range(y.shape[1])
]
else:
assert model.get_feature_names() == ["RandomForestClassifier__1"]
### Test with strings
if multi_output:
y = np.array(["a", "b", "c"])[np.random.randint(0, 3, 200).reshape(
(100, 2))]
else:
y = np.array(["a", "b", "c"])[np.random.randint(0, 3, 100)]
model = OutSamplerTransformer(
RandomForestClassifier(n_estimators=10, random_state=123))
model.fit(X, y)
p1 = model.model.predict_proba(X)
p2 = model.transform(X)
if multi_output:
assert isinstance(p1, list)
assert len(p1) == y.shape[1]
assert p2.shape == (100, 6)
assert np.abs(p1[0] - p2[:, 0:3]).max() <= 10**(-10)
assert np.abs(p1[1] - p2[:, 3:]).max() <= 10**(-10)
else:
assert p1.shape == (100, 3)
assert p2.shape == (100, 3)
assert np.abs(p1 - p2).max() <= 10**(-10)
assert model.get_feature_names() == [
"RandomForestClassifier__a",
"RandomForestClassifier__b",
"RandomForestClassifier__c",
]