将所选列的重复值替换为空字符串csv

Question

我遇到了一个可能非常具体的问题，我有以下数据:

输入

<表类>id名称phone1phone2phone3tbody><<tr>1约翰3222430085998314692412345678902瓦妮莎322243008599342542433玛丽亚7861628239786162823978616282394奥斯卡786115788832224300854保罗715108791255555555555保罗71510879125555555555

Filter,phone，然后stack，重塑成一个多索引序列，然后删除重复的序列，然后unstack,assign，结果返回到原始数据帧

df.assign(**df.filter(like='phone').stack().drop_duplicates().unstack())

---

id     name      phone1      phone2      phone3
0  1     John  3222430085  9983146924  1234567890
1  2  Vanessa         NaN  9934254243         NaN
2  3    Maria  7861628239         NaN         NaN
3  4    Oscar  7861157888         NaN         NaN
4  4     Paul  7151087912  5555555555         NaN
5  5     Paul         NaN         NaN         NaN