基于梯度提升树实现波士顿房价预测

基于梯度提升树实现波士顿房价预测

将波士顿房价数据集拆分成训练集和测试集，搭建gbdt_1、gbdt_2和gbdt_3三个梯度提升树模型，分布设置超参数n_estimators为50、100、150。各自对训练集进行训练，然后分别对训练集和测试集进行预测。输出以下结果：

（1）gbdt_1（50）在训练集上的准确率，在测试集上的准确率。

（2）gbdt_2（100）在训练集上的准确率，在测试集上的准确率。

（2）gbdt_3（150）在训练集上的准确率，在测试集上的准确率。

源码

from sklearn.datasets import load_boston
from sklearn.ensemble import GradientBoostingRegressor as GBDT
from sklearn.model_selection import train_test_split, validation_curve

def train():
    boston = load_boston()
    x_train, x_test, y_train, y_test = train_test_split(boston.data, boston.target)
    model = GBDT(n_estimators=50)
    model.fit(x_train, y_train)
    train_score = model.score(x_train, y_train)
    test_score = model.score(x_test, y_test)
    print("李思强 20201107148")
    print("50")
    print("train_score",train_score)
    print("test_score",test_score)

    mode2 = GBDT(n_estimators=100)
    mode2.fit(x_train, y_train)
    train_score2 = mode2.score(x_train, y_train)
    test_score2 = mode2.score(x_test, y_test)
    print("100")
    print("train_score",train_score2)
    print("test_score",test_score2)

    mode3 = GBDT(n_estimators=150)
    mode3.fit(x_train, y_train)
    train_score3 = mode3.score(x_train, y_train)
    test_score3 = mode3.score(x_test, y_test)
    print("150")
    print("train_score",train_score3)
    print("test_score",test_score3)
if __name__ == '__main__':
    train()

运行结果