Python3 多元回归（包含属性的向量化）

 # -*- coding: utf-8 -*-
 """
 Created on Thu Jan  4 19:52:03 2018

 @author: markli
 """
 import numpy as np;
 import pandas as pd;
 from sklearn import linear_model;
 from sklearn.externals import joblib;
 import sys;

 sys.path.append('../');
 path = 'DeliverIncludeClass.csv';
 #names=['Miles','NumberDeliveries','TravelTime'];
 reader = pd.read_csv(path,header=0,names=['Number','Miles','NumberDeliveries','Car','TravelTime']);

 #print(reader);
 X = np.array(reader[['Miles','NumberDeliveries']]);
 Y = np.array(reader['TravelTime']);
 Car = np.array(reader['Car']); #将为类别属性的数据列单独读出来
 #print(X);
 #print(Y);
 car_feature = list(set(Car));
 VectoreCar = np.zeros((len(Car),len(car_feature)));
 #本例中Car属性为类别属性，将其向量化
 for i in range(len(Car)):
     VectoreCar[i][Car[i]]=1;

 print(VectoreCar);
 print(Car);

 #将向量化的类别属性值与其他数据合并
 X = np.hstack((X,VectoreCar));
 print(X);

 reg = linear_model.LinearRegression();
 reg.fit(X,Y);
 joblib.dump(reg,'IncludeClassAttrRegression.pkl');
 print("系数 %s " %(reg.coef_));
 print("常系数 %s" %(reg.intercept_) );
 #
 x_test = [[80,5,0,1,0]];
 y_predict = reg.predict(x_test);
 print(y_predict);

DeliverIncludeClass.csv 文件格式