三、sklearn實現梯度下降

1、梯度下降的概念

　　https://1drv.ms/u/s!AlpiCfg9Lzmug1Dj69iU4OC0Rpt-

2、梯度下降的類型

• 批量梯度下降
• 隨機梯度下降
• 小批量梯度下降

3、sklearn實現梯度下降

# 隨機梯度下降
from sklearn.linear_model import SGDRegressor

from sklearn.preprocessing import StandardScaler

# 歸一化數據
std = StandardScaler()
std.fit(X_train)
X_train_std = std.transform(X_train)
X_test_std = std.transform(X_test)

# n_iter代表瀏覽多少次，默認是5
sgd_reg = SGDRegressor(n_iter=100)
sgd_reg.fit(X_train_std, y_train)
sgd_reg.score(X_test_std, y_test)