数据清洗

1、读取本地数据库 

import pymysql
import numpy as npy
import pandas as pda
import matplotlib.pylab as pyl
conn = pymysql.connect(host="localhost",user="root",passwd="360421",db="employees")   #读取本地数据库
sql = "select * from salaries limit 0,10000"                                                                              #SQL语句选择要读取的数据
data=pda.read_sql(sql,conn)
print(data.describe)
len(data)      #对比看有没有缺失
#发现缺失值
data2=data.T
salary=data2.values[1]
from_date=data2.values[2]
pyl.plot(from_date,salary,'o')
pyl.show()

2、读取结果未发现异常值 

 3、数据发现异常数据

   准备csv文件设置好缺失值和异常值

data=pda.read_csv("E:\AWorkStation01\数据挖掘\square.csv")    #读取
data1=data.T   #转置
row=data1.values[0]
col=data1.values[1]
pyl.plot(row,col,'o')
pyl.show()

结果

清洗异常数据

 

data1=data.values                                          #不能直接修改data的值
for i in range(len(data1)):                                #找出不满足条件的数据并修正
	if(data1[i][1]!=(i+1)*(i+1)):
		data1[i][1]=(i+1)*(i+1)
data2=data1.T
x=data2[0]
y=data2[1]
pyl.plot(x,y,'o')
pyl.show()