sklearn

一、Logisitic Regression模型

1. 取数据

import pandas as pd

df = pd.read_csv('/Users/cy_ariel/Downloads/adultTest.csv')

2. 数据处理

查看class字段取值

df['class'].value_counts()

<=50K    24720
 >50K      7841
Name: class, dtype: int64

更改class为0、1并去除

df.loc[df['class'] == "<=50K",'target'] = 0
df.loc[df['class'] != "<=50K",'target'] = 1

查看数据类型

df.dtypes

将object类型转为int，并去除class列

df1 = pd.get_dummies(data=df,columns=['workclass','education','marital-status','occupation','relationship','race','sex','native-country'])

df1.drop("class",inplace=True,axis=1)

3. 获取训练集测试集

获取X、Y

xdata = df1.drop('target',axis=1)
ydata = df1['target']

区分训练集和数据集

from sklearn.cross_validation import train_test_split

xtrain,xtest,ytrain,ytest = train_test_split(xdata,ydata,test_size=0.3)

4. 建模

建LR模型

from sklearn.linear_model import LogisticRegression

lr = LogisticRegression(C=2.0,max_iter=1)

训练模型

lr.fit(xtrain,ytrain)

应用模型

lr.predict(xtest)

模型得分

lr.score(xtest,ytest)