每个不同参与者的相应周列以及开始日期

Question

我有一个示例数据框，如下所示。

import pandas as pd
data = {'ID':['A','A','A','A','A', 'B','B','B','C','C','C','C'],
    'Date':['9/20/2021  4:34:57 AM', '9/21/2021  6:54:22 PM', '9/28/2021  5:54:22 PM', 
           '9/30/2021  6:54:22 PM', '10/5/2021  6:54:22 PM',
           '9/4/2021  7:04:38 PM','9/9/2021  7:04:38 PM','9/17/2021  7:04:38 PM',
           '9/28/2021  3:21:23 AM','9/30/2021  3:21:23 AM','10/05/2021  3:21:23 
            AM','10/15/2021  3:21:23 AM']}
    
 df1 = pd.DataFrame(data)
 df1

'Date' 列采用日期时间格式。

现在对于每个参与者，第一个条目是注册日期，这将是该特定参与者第一周的开始。我想要后续日期条目的相应周数，将第一个条目视为第 0 天。

最终的数据帧应该如下图所示。

非常感谢任何帮助。谢谢。

Answer 1

尝试：

df['Date'] = pd.to_datetime(df['Date'])

df['Week'] = df.groupby('ID')['Date'] \
               .apply(lambda t: 1 + (t - min(t)).dt.days // 7)

df['Week'] = 'Week' + df['Week'].astype(str)

输出：

>>> df
   ID                Date   Week
0   A 2021-09-20 04:34:57  Week1
1   A 2021-09-21 18:54:22  Week1
2   A 2021-09-28 17:54:22  Week2
3   A 2021-09-30 18:54:22  Week2
4   A 2021-10-05 18:54:22  Week3
5   B 2021-09-04 19:04:38  Week1
6   B 2021-09-09 19:04:38  Week1
7   B 2021-09-17 19:04:38  Week2
8   C 2021-09-28 03:21:23  Week1
9   C 2021-09-30 03:21:23  Week1
10  C 2021-10-05 03:21:23  Week2
11  C 2021-10-15 03:21:23  Week3

由@ddejohn修复

Answer 2

您可以使用：

df1['Date'] = pd.to_datetime(df1['Date'])

df1['Week'] = (df1.groupby('ID')['Date'].diff().dt.days
                  .groupby(df1['ID']).cumsum()
                  .fillna(0, downcast='infer') // 7 + 1)

df1['Week'] = 'Week' + df1['Week'].astype(str)

结果：

print(df1)

   ID                Date   Week
0   A 2021-09-20 04:34:57  Week1
1   A 2021-09-21 18:54:22  Week1
2   A 2021-09-28 17:54:22  Week2
3   A 2021-09-30 18:54:22  Week2
4   A 2021-10-05 18:54:22  Week3
5   B 2021-09-04 19:04:38  Week1
6   B 2021-09-09 19:04:38  Week1
7   B 2021-09-17 19:04:38  Week2
8   C 2021-09-28 03:21:23  Week1
9   C 2021-09-30 03:21:23  Week1
10  C 2021-10-05 03:21:23  Week2
11  C 2021-10-15 03:21:23  Week3

每个不同参与者的相应周列以及开始日期

Corresponding week column for each different participant with starting dates

python

dataframe

pandas

data-science

data-preprocessing