#from sklearn.datasets import make_friedman3
#X,y = make_friedman3(n_samples=800,noise=0.0111,random_state=None)
#使用Bagging和IterativeBagging算法进行预测并输出均方差
from sklearn.ensemble import BaggingRegressor
from sklearn.metrics import mean_squared_error
#使用Bagging算法进行回归预测
br = BaggingRegressor(n_estimators=80, oob_score=True)
br.fit(X, y)
print("BaggingRegressor:train")
#包内回归测试
predict_train = br.predict(X)
print(mean_squared_error(y, predict_train))
#包外回归测试
predict_train = br._lpz_predict(X, y)
print(mean_squared_error(y, predict_train))
#print(bc.oob_score_)
#print("BaggingRegressor:test")
#predict = br.predict(X_test)
#print(mean_squared_error(y_test,predict))
y1 = y
err = mean_squared_error(y, predict_train)
min_err = err
#使用IterativeBagging算法进行回归预测
print("IterativeBagging")
for i in range(1):
#predict test data
y1 = y1 - br._lpz_predict(X, y1)
br.fit(X, y1)
