Simple and stupid benchmark for sklearn DecisionTreeRegressor

bench_tree.py

import numpy as np
from sklearn import datasets
from sklearn.ensemble import gradient_boosting
from sklearn.ensemble import RandomForestClassifier,RandomForestRegressor
from sklearn.tree import DecisionTreeClassifier, DecisionTreeRegressor

X, y = datasets.make_hastie_10_2(n_samples=12000, random_state=1)
X = X.astype(np.float32)


print "%timeit clf.fit(X, y)"

boston = datasets.load_boston()

X, y = boston.data, boston.target


clf = DecisionTreeRegressor(max_depth=20)
%timeit clf.fit(X, y)
%timeit clf.predict(X)

clf = DecisionTreeRegressor(max_depth=1)
%timeit clf.fit(X, y)
%timeit clf.predict(X)

clf = RandomForestRegressor()
%timeit clf.fit(X, y)
%timeit clf.predict(X)


clf = gradient_boosting.GradientBoostingRegressor(n_estimators=250,
                                                  max_depth=1,
                                                  learn_rate=1.0)
%timeit clf.fit(X, y)
%timeit clf.predict(X)