python merge on多个条件_【链接】pandas中的merge应用

数据分析方面，“连接”是高频出现的一个词。在python应用种，merge方法就是一种重要的连接方法。

merge主要功能在于合并数据集，通过left、right来连结字段。

1.merge默认按相同字段名合并，且取两个都有的。

import pandas  as pd
df1 = pd.DataFrame({"name":["kate","herz","catherine","sally"],
                    "age":[25,28,39,35]})
df2 = pd.DataFrame({"name":["kate","herz","sally"],
                     "score":[70,60,90]})
pd.merge(df1,df2)

2.当左右连结字段名不相同时，使用left_on,right_on

df1 = pd.DataFrame({"name":["kate","herz","catherine","sally"],
                    "age":[25,28,39,35]})
df2 = pd.DataFrame({"call_name":["kate","herz","sally"],
                     "score":[70,60,90]})
pd.merge(df1,df2,left_on = "name",right_on = "call_name")

3.合并后，删除重复的列

df1 = pd.DataFrame({"name":["kate","herz","catherine","sally"],
                    "age":[25,28,39,35]})
df2 = pd.DataFrame({"call_name":["kate","herz","sally"],
                     "score":[70,60,90]})
pd.merge(df1,df2,left_on = "name",right_on = "call_name").drop("name",axis = 1)

4.参数how的使用

• inner 内连接，取交集

pd.merge(df1,df2,left_on = "name",right_on = "call_name",how ="inner")

• outer 外连接，取并集，并用nan填充

df3 = pd.DataFrame({"name":["kate","herz","sally","cristin"],"score":[70,60,90,30]})
pd.merge(df1,df3,on="name",how="outer")

• left 左连接，左侧取全部，右侧取部分

pd.merge(df1,df3,on = "name",how = "left")

• right 右链接，左侧取部分，右侧取全部

pd.merge(df1,df3,on = "name",how = "right")