删除Pandas中的重复项,不包括一列

Question

这看起来很简单,但我在互联网上找不到任何相关信息

我有一个如下的数据框

City    State Zip           Date        Description       
Earlham IA    50072-1036    2014-10-10  Postmarket Assurance: Devices
Earlham IA    50072-1036    2014-10-10  Compliance: Devices
Madrid  IA    50156-1748    2014-09-10  Drug Quality Assurance

如何消除与5列中的4列匹配的重复项？列不匹配Description.

结果将是

City    State Zip           Date        Description       
Earlham IA    50072-1036    2014-10-10  Postmarket Assurance: Devices
Madrid  IA    50156-1748    2014-09-10  Drug Quality Assurance

我在网上找到的是drop_dupilcates与subset参数可以工作,但我不确定我怎么可以把它应用到多个列.

Answer 1

你实际上找到了解决方案.对于多列,子集将是一个列表.

df.drop_duplicates(subset=['City', 'State', 'Zip', 'Date']) 

或者,只需声明要忽略的列:

df.drop_duplicates(subset=df.columns.difference(['Description']))