How to predict class labels in X and their probability estimates?

[andresilvapimentel]

How can I predict class labels in X and their probability estimates using superlearner with the mlens library?
I tried to use:

from mlens.ensemble import SuperLearner
from mlens.ensemble.base import BaseEnsemble

make predictions on hold out set

yhat = ensemble.predict(X_val, proba=True)
print('Super Learner: %.3f' % (accuracy_score(y_val, yhat) * 100))
yhat_proba = ensemble.predict_proba(X_val, proba=True)
print(yhat_proba)

but it did not work at al.