机器学习基础之《回归与聚类算法（6）—模型保存与加载》

算法（6）—模型保存与加载》"/>

机器学习基础之《回归与聚类算法（6）—模型保存与加载》

一、背景

现在我们预测每次都要重新运行一遍模型。完整的流程应该是不断调整阈值重复计算。
当训练或者计算好一个模型之后，那么如果别人需要我们提供结果预测，就需要保存模型（主要是保存算法的参数）。

二、sklearn模型的保存和加载API

1、import joblib
保存：joblib.dump(rf, "test.pkl")
    rf：是预估器estimator
    test.pkl：是保存的名字
    将预估器序列化保存在本地    
加载：estimator = joblib.load("test.pkl")

2、代码

from sklearn.datasets import load_boston
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.linear_model import LinearRegression, SGDRegressor, Ridge
from sklearn.metrics import mean_squared_error
import joblibdef linear1():"""正规方程的优化方法对波士顿房价进行预测"""# 1、获取数据boston = load_boston()# 2、划分数据集x_train,x_test, y_train, y_test = train_test_split(boston.data, boston.target, random_state=10)# 3、标准化transfer = StandardScaler()x_train = transfer.fit_transform(x_train)x_test = transfer.transform(x_test)# 4、预估器estimator = LinearRegression()estimator.fit(x_train, y_train)# 5、得出模型print("正规方程-权重系数为：\n", estimator.coef_)print("正规方程-偏置为：\n", estimator.intercept_)# 6、模型评估y_predict = estimator.predict(x_test)print("预测房价：\n", y_predict)error = mean_squared_error(y_test, y_predict)print("正规方程-均方误差为：\n", error)return Nonedef linear2():"""梯度下降的优化方法对波士顿房价进行预测"""# 1、获取数据boston = load_boston()# 2、划分数据集x_train,x_test, y_train, y_test = train_test_split(boston.data, boston.target, random_state=10)# 3、标准化transfer = StandardScaler()x_train = transfer.fit_transform(x_train)x_test = transfer.transform(x_test)# 4、预估器estimator = SGDRegressor()estimator.fit(x_train, y_train)# 5、得出模型print("梯度下降-权重系数为：\n", estimator.coef_)print("梯度下降-偏置为：\n", estimator.intercept_)# 6、模型评估y_predict = estimator.predict(x_test)print("预测房价：\n", y_predict)error = mean_squared_error(y_test, y_predict)print("梯度下降-均方误差为：\n", error)return Nonedef linear3():"""岭回归对波士顿房价进行预测"""# 1、获取数据boston = load_boston()# 2、划分数据集x_train,x_test, y_train, y_test = train_test_split(boston.data, boston.target, random_state=10)# 3、标准化transfer = StandardScaler()x_train = transfer.fit_transform(x_train)x_test = transfer.transform(x_test)# 4、预估器estimator = Ridge()estimator.fit(x_train, y_train)# 保存模型joblib.dump(estimator, "my_ridge.pkl")# 5、得出模型print("岭回归-权重系数为：\n", estimator.coef_)print("岭回归-偏置为：\n", estimator.intercept_)# 6、模型评估y_predict = estimator.predict(x_test)print("预测房价：\n", y_predict)error = mean_squared_error(y_test, y_predict)print("岭回归-均方误差为：\n", error)return Nonedef linear4():"""岭回归对波士顿房价进行预测"""# 1、获取数据boston = load_boston()# 2、划分数据集x_train,x_test, y_train, y_test = train_test_split(boston.data, boston.target, random_state=10)# 3、标准化transfer = StandardScaler()x_train = transfer.fit_transform(x_train)x_test = transfer.transform(x_test)# 加载模型estimator = joblib.load("my_ridge.pkl")# 5、得出模型print("岭回归-权重系数为：\n", estimator.coef_)print("岭回归-偏置为：\n", estimator.intercept_)# 6、模型评估y_predict = estimator.predict(x_test)print("预测房价：\n", y_predict)error = mean_squared_error(y_test, y_predict)print("岭回归-均方误差为：\n", error)return Noneif __name__ == "__main__":# 代码1：正规方程的优化方法对波士顿房价进行预测linear1()# 代码2：梯度下降的优化方法对波士顿房价进行预测linear2()# 代码3：岭回归对波士顿房价进行预测linear3()# 代码4：加载模型linear4()