线性回归之模型的保存和加载

1 sklearn模型的保存和加载API

• from sklearn.externals import joblib   【目前这行代码报错，直接写import joblib就可以了】
  • 保存：joblib.dump(estimator, 'test.pkl')
  • 加载：estimator = joblib.load('test.pkl')
  • 【注意：1.保存文件，后缀名是**.pkl；2.加载模型是需要通过一个变量进行承接】

2 线性回归的模型保存加载案例

from sklearn.datasets import load_boston
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import mean_squared_error
from sklearn.linear_model import Ridge, RidgeCV
import joblibdef linear_model_demo():"""线性回归：岭回归:return:"""#  1.获取数据data = load_boston()#  2.数据集划分x_train, x_test, y_train, y_test = train_test_split(data.data, data.target, random_state=22)#  3.特征工程-标准化transter = StandardScaler()x_train = transter.fit_transform(x_train)x_test = transter.fit_transform(x_test)#  4.机器学习-线性回归（岭回归）# #  4.1模型训练# estimator = Ridge(alpha=1)# # estimator = RidgeCV(alphas=(0.1, 1, 10))# estimator.fit(x_train, y_train)# #  4.2模型保存# joblib.dump(estimator, "./test.pkl")#  4.3加载模型estimator = joblib.load("./test.pkl")#  5.模型评估#  5.1获取系数等值y_predict = estimator.predict(x_test)print("预测值为：\n", y_predict)print("模型中的系数为:\n", estimator.coef_)print("模型中的偏执为：\n", estimator.intercept_)#  5.2评价#  均方误差error = mean_squared_error(y_test, y_predict)print("误差为：\n", error)linear_model_demo()

运行结果：

注意：