如何在以下数据集上使用groupby

Question

我有这个数据集，

sender          team_id     receiver
John Cena         1           Margaret
Genghis Khan      2           Mahathma
Mahathma Gandhi   1           John
John Doe          2           Genghis
Margaret Thatcher 1           John

每个发送者都有一个团队ID，接收者的名字只是他们的名字。我想找出每个消息是否在团队成员之间。结果看起来像这样。

sender          team_id     receiver       btwn_teammates
John Cena         1           Margaret          Yes
Genghis Khan      2           Mahathma          No
Mahathma Gandhi   1           John              Yes
John Doe          2           Genghis           Yes
Margaret Thatcher 1           John              Yes

Answer 1

合并名称+ team_id的第一部分，然后映射指标值：

df2 = df[['sender', 'team_id']].rename(columns={'sender': 'receiver'})
df2['receiver'] = df2.receiver.str.split().str[0]
df2 = df2.drop_duplicates()  # So left merge preserves size.

df = df.merge(df2, how='left', indicator='btwn_team')
df['btwn_team'] = df.btwn_team.map({'both': 'Yes', 'left_only': 'No'})

输出：

              sender  team_id  receiver btwn_team
0          John Cena        1  Margaret       Yes
1       Genghis Khan        2  Mahathma        No
2    Mahathma Gandhi        1      John       Yes
3           John Doe        2   Genghis       Yes
4  Margaret Thatcher        1      John       Yes