逻辑回归预测收入

# 导入库：KNeighborsClassifier
import numpy as np
from sklearn.neighbors import KNeighborsClassifier
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap
%matplotlib inline
# 导入sklearn自带数据集
from sklearn import datasets
import  pandas as  pd
from  sklearn.linear_model import  LogisticRegression 

path=r"C:\Users\Tsinghua-yincheng\Desktop\SZday93\adults.txt"
data=pd.read_csv(path)
print(type(data))
data["age"]

data.shape
data.index
data.columns
X=data[['age','education','marital_status','sex' ,'hours_per_week']]
X

Y=data['salary']
Y

X=data[['age','education','occupation' ,'hours_per_week']]
X

X["education"].unique().size
X["occupation"].unique()
def  edu2float(edu): #任何学历，返回一个数字代表
    return  np.argwhere(X["education"].unique()==edu)[0,0]
def  occupation2float(occupation): #任何学历，返回一个数字代表
    return  np.argwhere(X["occupation"].unique()==occupation)[0,0]
X["occupation"]=X["occupation"].map(occupation2float)
X["education"]=X["education"].map(edu2float)

X_train=X.iloc[:-1000] #训练数据
X_test=X.iloc[-1000:]#测试数据
Y_train=Y.iloc[:-1000] #训练数据
Y_test=Y.iloc[-1000:]#测试数据
logic=LogisticRegression()
logic.fit(X_train,Y_train)
Ynew=logic.predict(X_test) #用测试数据产生预测结果
result=(Y_test==Ynew) #统计相等个数
accuracy=result.sum()/1000 #统计成功率
accuracy