def test_fit_predict_standard_stack(teardown):
# This uses the "standard" protocol where the 2nd level features
# are the out-of-fold predictions of the 1st. It also appends the
# original data to the 2nd level features.
# See for example: https://www.kdnuggets.com/2017/02/stacking-models-imropved-predictions.html
X_data, y_t_data = breast_cancer.data, breast_cancer.target
X_train, X_test, y_t_train, y_t_test = train_test_split(X_data,
y_t_data,
test_size=0.2,
random_state=0)
random_state = 42
# baikal way
x = Input()
y_t = Input()
y_p1 = RandomForestClassifierOOF(n_estimators=10,
random_state=random_state)(
x, y_t, compute_func="predict_proba")
y_p1 = Lambda(lambda array: array[:, 1:])(y_p1) # remove collinear feature
x_scaled = StandardScaler()(x)
y_p2 = LinearSVCOOF(random_state=random_state)(
x_scaled, y_t, compute_func="decision_function")
stacked_features = ColumnStack()([x, y_p1, y_p2])
y_p = LogisticRegression(solver="liblinear",
random_state=random_state)(stacked_features, y_t)
model = Model(x, y_p, y_t)
model.fit(X_train, y_t_train)
y_pred_baikal = model.predict(X_test)
# traditional way
estimators = [
("rf",
RandomForestClassifier(n_estimators=10, random_state=random_state)),
("svr",
make_pipeline(StandardScaler(),
LinearSVC(random_state=random_state))),
]
clf = sklearn.ensemble.StackingClassifier(
estimators=estimators,
final_estimator=LogisticRegression(solver="liblinear",
random_state=random_state),
passthrough=True,
)
y_pred_traditional = clf.fit(X_train, y_t_train).predict(X_test)
assert_array_equal(y_pred_baikal, y_pred_traditional)
def dataplaceholders(self):
x1 = Input(name="x1")
x2 = Input(name="x2")
y1_t = Input(name="y1_t")
x1_rescaled = StandardScaler()(x1)
y1 = LogisticRegression()(x1_rescaled, y1_t)
y2 = PCA()(x2)
return x1, x2, x1_rescaled, y1, y2, y1_t
def test_fit_with_shared_step(self, teardown):
x = Input()
scaler = StandardScaler()
z = scaler(x, compute_func="transform", trainable=True)
y = scaler(z, compute_func="inverse_transform", trainable=False)
model = Model(x, y)
model.fit(np.array([1, 3, 1, 3]).reshape(-1, 1))
assert (scaler.mean_, scaler.var_) == (2.0, 1.0)
def test_fit_predict_with_shared_step(teardown):
x = Input()
scaler = StandardScaler()
z = scaler(x, compute_func="transform", trainable=True)
y = scaler(z, compute_func="inverse_transform", trainable=False)
model = Model(x, y)
X_data = np.array([1, 3, 1, 3]).reshape(-1, 1)
model.fit(X_data)
assert_array_equal(model.predict(X_data), X_data)
def test_get_set_params_invariance(teardown):
scaler = StandardScaler(name="scaler")
regressor = LinearRegression(name="regressor")
x = Input()
y_t = Input()
y_t_scaled = scaler(y_t)
y_p_scaled = regressor(x, y_t_scaled)
y_p = scaler(y_p_scaled, compute_func="inverse_transform", trainable=False)
model = Model(x, y_p, y_t)
params1 = model.get_params()
model.set_params(**params1)
params2 = model.get_params()
assert params2 == params1