how to calculate time difference for each day for every employee in pandas?-CodePudding

i have this dataframe

    Matricule         DateTime        Date      Time
1   1000    2022-01-06 10:59:51 2022-01-06  10:59:51
2   1000    2022-01-07 08:40:09 2022-01-07  08:40:09
3   1000    2022-01-26 15:39:10 2022-01-26  15:39:10
4   1104    2022-01-03 14:33:38 2022-01-03  14:33:38
81  1104    2022-01-04 10:04:18 2022-01-04  10:04:18
... ... ... ... ...
1548    1803    2022-01-24 15:51:22 2022-01-24  15:51:22
1549    1803    2022-01-24 15:51:29 2022-01-24  15:51:29
1550    1803    2022-01-24 16:54:23 2022-01-24  16:54:23
1551    1803    2022-01-28 14:42:01 2022-01-28  14:42:01
1552    1803    2022-01-28 14:42:32 2022-01-28  14:42:32

i want to calculate time difference between the first time of the day and last time of the day for each day for every employee to know how much hours he spent at work daily for exemple

Matricule          Date  WorkTime      
1   1000     2022-01-06  1
2   1000     2022-01-07  3
3   1000     2022-01-26  5
4   1104     2022-01-03  2
81  1104     2022-01-04  8

CodePudding user response：

you can use the split-apply-combine approach, write a func for each group and apply on the groupby:

grpd = df.groupby(['Matricule', 'Date'])

def get_hours(df):
    start = df['Time'].min()
    end = df['Time'].max()
    new_df = pd.DataFrame([end-start], columns=['WorkTime'])
    return new_df

grpd.apply(get_hours)