python数据分析基础-表拼接

linux+pycharm+anaconda

#多表拼接,merge默认用两列公共列为链接键,on来指定链接键,可为多列
import pandas as pd
import numpy as np
 data1=pd.DataFrame([[20,99999,6,"c"],[30,999999,6,"c++"],[20,999994,8,"UI"],[27,999999,6,"c++"],[30,999999,6,"python"],[27,999999,7,"python"]])
 data1.columns=["年龄","收入","家属数","岗位"]
data2=pd.DataFrame([["zs",321,"c"],["ls",421,"c++"],["ww",994,"UI"],["zl",211,"c++"],["tq",943,"python"],["wb",876,"python"]],columns=["姓名","工号","岗位"])

#一对一,横向
new_data=pd.merge(data1,data2)


#多对一,横向on指定链接键
pd.merge(data1,data2,on="工号")

#多对多
pd.merge(data2,data1)

#多表链接,left_on,right_on,链接left_on==right_on的所有列
pd.merge(data1,data2,left_on="岗位",right_on='工号')

#索引链接left_index和right_index
pd.merge(data2,data1,left_index=True,right_index=True)

#索引列和普通列混用
pd.merge(data1,data2,left_index=True,right_on="工号")

#inner内链接,不指明都为内连接
pd.merge(data1,data2,on="岗位",how="inner")

#左链接以左表为基础,右链接同理,找不到用nan填充
pd.merge(data1,data2,on="岗位",how="left")
pd.merge(data1,data2,on="岗位",how="right")

#外链接,取两表并集,重复列自动加后缀,默认为["_x","_y"],suffixes=["_L","_R"]可改名
pd.merge(data2,data1,on="岗位",how="outer")


#纵向拼接concat(),ignore_index=True重置索引,drop_duplicates()删除重复值
new_data=pd.concat([data1,data2],ignore_index=True).drop_duplicates()