[bagging and pasting] #sklearn #machinelearning #classification #ensemble

bagging-and-pasting.py

from sklearn.ensemble import BaggingClassifier # or BaggingRegressor for regression
from sklearn.tree import DecisionTreeClassifier

# set bootstrap to False if you want to do pasting (sampling without replacement)
# n_jobs sets how many cores to use (-1 means all)
# oob_score is the out-of-bag score
bag_clf = BaggingClassifier(DecisionTreeClassifier(),
                            n_estimators=500,
                            max_samples=100,
                            bootstrap=True,
                            n_jobs=-1,
                            bootstrap_features=True,
                            max_features=4,
                            oob_score=True)

bag_clf.fit(X_train, y_train)
y_pred = bag_clf.predict(X_test)

# out-of-bag score
bag_clf.oob_score_