Python与数据挖掘(二)——逻辑回归

2、算法实现

import pandas as pd
from sklearn.linear_model import LogisticRegression, RandomizedLogisticRegreesion
from sklearn.cross_validation import train_test_split

#读入EXCEL文件，声明编码方式，默认方式是ASCII
data=pd.read_csv('E/Python/LogisticRegression.csv',encoding='utf-8')
将类别型变量进行独热编码，使字符串变为数字“0”，“1”
data_dum=pd.get_dummies(data,prefix='rank',columns=['rank'],drop_first=True)

#切分训练集和测试集
X_train,X_test, y_train,y_test=train_test_split(data_dum.ix[:, 1:],data_dum.ix[:,0],test_size=.1,random_state=520)

lr=LogisticRegression()
lr.fit(X_train,y_train)