我有一个像这样的pandas dataframe:
CustomerId Timestamp
0. a 01-09-2018 00:08:00
1. a 01-09-2018 00:09:00
2. b 01-09-2018 00:11:00
3. b 01-09-2018 00:15:00
我需要计算每个客户连续时间戳之间分钟的差异,以便最终我有一些看起来像这样的东西:
CustomerId Timestamp Difference
0. a 01-09-2018 00:08:00 -
1. a 01-09-2018 00:09:00 1
2. b 01-09-2018 00:11:00 -
3. b 01-09-2018 00:15:00 4
我一直在尝试一些循环,但似乎没有任何努力。如果有人可以帮助我,我真的很感激:)
使用 groupby
与 diff
df.groupby('CustomerId').Timestamp.diff().dt.total_seconds()/60
Out[10]:
0.0 NaN
1.0 1.0
2.0 NaN
3.0 4.0
Name: Timestamp, dtype: float64
df['Different']=df.groupby('CustomerId').Timestamp.diff().dt.total_seconds()/60