def test_proba_classif_convergence():
X_train, _, y_train, _ = load_scaled_boston()
y_train = np.round(y_train)
mc = MondrianTreeClassifier(random_state=0)
mc.fit(X_train, y_train)
lb = LabelBinarizer()
y_bin = lb.fit_transform(y_train)
le = LabelEncoder()
y_enc = le.fit_transform(y_train)
proba = mc.predict_proba(X_train)
labels = mc.predict(X_train)
assert_array_equal(proba, y_bin)
assert_array_equal(labels, lb.inverse_transform(y_bin))
# For points completely far away from the training data, this
# should converge to the empirical distribution of labels.
# X is scaled between to -1.0 and 1.0
X_inf = np.vstack(
(30.0 * np.ones(X_train.shape[1]), -30.0 * np.ones(X_train.shape[1])))
inf_proba = mc.predict_proba(X_inf)
emp_proba = np.bincount(y_enc) / float(len(y_enc))
assert_array_almost_equal(inf_proba, [emp_proba, emp_proba])
def test_proba_classif_convergence():
X_train, _, y_train, _ = load_scaled_boston()
y_train = np.round(y_train)
mc = MondrianTreeClassifier(random_state=0)
mc.fit(X_train, y_train)
check_proba_classif_convergence(X_train, y_train, mc)
mc.partial_fit(X_train, y_train)
check_proba_classif_convergence(X_train, y_train, mc)
def test_weighted_decision_path_classif():
X_train, X_test, y_train, y_test = load_scaled_boston()
y_train = np.round(y_train)
y_test = np.round(y_test)
mtc = MondrianTreeClassifier(random_state=0)
mtc.fit(X_train, np.round(y_train))
check_weighted_decision_path_classif(mtc, X_test)
mtc.partial_fit(X_train, np.round(y_train))
check_weighted_decision_path_classif(mtc, X_test)
def test_weighted_decision_path_test_classif():
X_train, X_test, y_train, y_test = load_scaled_boston()
y_train = np.round(y_train)
y_test = np.round(y_test)
n_train = X_train.shape[0]
mtc = MondrianTreeClassifier(random_state=0)
mtc.fit(X_train, np.round(y_train))
weights = mtc.weighted_decision_path(X_test)
node_probas = (mtc.tree_.value[:, 0, :] /
np.expand_dims(mtc.tree_.n_node_samples, axis=1))
probas1 = []
for startptr, endptr in zip(weights.indptr[:-1], weights.indptr[1:]):
curr_nodes = weights.indices[startptr:endptr]
curr_weights = np.expand_dims(weights.data[startptr:endptr], axis=1)
curr_probas = node_probas[curr_nodes]
probas1.append(np.sum(curr_weights * curr_probas, axis=0))
probas2 = mtc.predict_proba(X_test)
assert_array_almost_equal(probas1, probas2, 5)
from sklearn.preprocessing import LabelEncoder
from sklearn.preprocessing import MinMaxScaler
from skgarden.utils.testing import assert_array_equal
from skgarden.utils.testing import assert_array_almost_equal
from skgarden.utils.testing import assert_almost_equal
from skgarden.utils.testing import assert_equal
from skgarden.utils.testing import assert_false
from skgarden.utils.testing import assert_less
from skgarden.utils.testing import assert_true
from skgarden.mondrian import MondrianTreeClassifier
from skgarden.mondrian import MondrianTreeRegressor
estimators = [
MondrianTreeRegressor(random_state=0),
MondrianTreeClassifier(random_state=0)
]
def test_tree_predict():
X = np.array([[-2, -1], [-1, -1], [-1, -2], [1, 1], [1, 2], [2, 1]])
y = [-1, -1, -1, 1, 1, 1]
T = [[-1, -1], [2, 2], [3, 2]]
# This test is dependent on the random-state since the feature
# and the threshold selected at every split is independent of the
# label.
for est_true in estimators:
est = clone(est_true)
est.set_params(random_state=0, max_depth=1)
est.fit(X, y)