def test_weighted_decision_path():
X_train, X_test, y_train, y_test = train_test_split(X,
y,
train_size=0.6,
test_size=0.4)
mr = MondrianForestRegressor(random_state=0)
mr.fit(X_train, y_train)
# decision_path is implemented in sklearn while
# weighted_decision_path is implemented here so check
paths, col_inds = mr.decision_path(X_train)
weight_paths, weight_col_inds = mr.weighted_decision_path(X_train)
assert_array_equal(col_inds, weight_col_inds)
n_nodes = [est.tree_.node_count for est in mr.estimators_]
assert_equal(weight_paths.shape[0], X_train.shape[0])
assert_equal(weight_paths.shape[1], sum(n_nodes))
# We are calculating the weighted decision path on train data, so
# the weights should be concentrated at the leaves.
leaf_indices = mr.apply(X_train)
for est_ind, curr_leaf_indices in enumerate(leaf_indices.T):
curr_path = weight_paths[:, col_inds[est_ind]:col_inds[est_ind +
1]].toarray()
assert_array_equal(np.where(curr_path)[1], curr_leaf_indices)
# Sum of the weights across all the nodes in each estimator
# for each sample should sum up to 1.0
assert_array_almost_equal(
np.ravel(mr.weighted_decision_path(X_test)[0].sum(axis=1)),
mr.n_estimators * np.ones(X_test.shape[0]), 5)
def test_decision_path():
mr = MondrianForestRegressor(random_state=0)
mr.fit(X, y)
indicator, col_inds = mr.decision_path(X)
indices, indptr, data = indicator.indices, indicator.indptr, indicator.data
n_nodes = [est.tree_.node_count for est in mr.estimators_]
assert_equal(indicator.shape[0], X.shape[0])
assert_equal(indicator.shape[1], sum(n_nodes))
assert_array_equal(np.diff(col_inds), n_nodes)
# Check that all leaf nodes are in the decision path.
leaf_indices = mr.apply(X) + np.reshape(col_inds[:-1], (1, -1))
for sample_ind, curr_leaf in enumerate(leaf_indices):
sample_indices = indices[indptr[sample_ind]:indptr[sample_ind + 1]]
assert_true(np.all(np.in1d(curr_leaf, sample_indices)))