How to set a custom day of week mapping
Question:
I am trying to create my own custom day of the week mapping using python. I have used a few different methods such as dayofweek and isoweekday. Yet all of these dont provide me with what I need.
0 2026-01-01 00:00:00
1 2026-01-01 01:00:00
2 2026-01-01 02:00:00
3 2026-01-01 03:00:00
4 2026-01-01 04:00:00
5 2026-01-01 05:00:00
6 2026-01-01 06:00:00
7 2026-01-01 07:00:00
8 2026-01-01 08:00:00
9 2026-01-01 09:00:00
10 2026-01-01 10:00:00
11 2026-01-01 11:00:00
12 2026-01-01 12:00:00
13 2026-01-01 13:00:00
14 2026-01-01 14:00:00
15 2026-01-01 15:00:00
16 2026-01-01 16:00:00
17 2026-01-01 17:00:00
18 2026-01-01 18:00:00
19 2026-01-01 19:00:00
Name: Date, dtype: datetime64[ns]
It contiunes to
0 2026-01-01 00:00:00
1 2026-01-01 01:00:00
2 2026-01-01 02:00:00
3 2026-01-01 03:00:00
4 2026-01-01 04:00:00
95 2026-01-04 23:00:00
96 2026-01-05 00:00:00
97 2026-01-05 01:00:00
98 2026-01-05 02:00:00
99 2026-01-05 03:00:00
example of my code
power_data['Day of Week'] = power_data['Date'].dt.dayofweek+1
power_data['Day of Weekv2'] = power_data['Date'].dt.isoweekday
Above is a example portion of my dataframe, the data formatting I would like to follow is Sunday be 1, monday be 2…etc and Saturday be equal to 7. Please let me know if I can do this how its currently presented
Answers:
As per pandas documentation for weekday, it mentions that:
The day of the week with Monday=0, Sunday=6.
Which means, you just need to add 2 to the weekday value as you want to shift Monday from 0 to 2, then take modulo by 8
# df is your dataframe, and date is the column name consisting pandas Timestamp
>>> (df['date'].dt.weekday+2)%8
#output:
0 5
1 5
2 5
3 5
4 5
5 5
6 5
7 5
8 5
9 5
10 5
11 5
12 5
13 5
14 5
15 5
16 5
17 5
18 5
Name: date, dtype: int64
I am trying to create my own custom day of the week mapping using python. I have used a few different methods such as dayofweek and isoweekday. Yet all of these dont provide me with what I need.
0 2026-01-01 00:00:00
1 2026-01-01 01:00:00
2 2026-01-01 02:00:00
3 2026-01-01 03:00:00
4 2026-01-01 04:00:00
5 2026-01-01 05:00:00
6 2026-01-01 06:00:00
7 2026-01-01 07:00:00
8 2026-01-01 08:00:00
9 2026-01-01 09:00:00
10 2026-01-01 10:00:00
11 2026-01-01 11:00:00
12 2026-01-01 12:00:00
13 2026-01-01 13:00:00
14 2026-01-01 14:00:00
15 2026-01-01 15:00:00
16 2026-01-01 16:00:00
17 2026-01-01 17:00:00
18 2026-01-01 18:00:00
19 2026-01-01 19:00:00
Name: Date, dtype: datetime64[ns]
It contiunes to
0 2026-01-01 00:00:00
1 2026-01-01 01:00:00
2 2026-01-01 02:00:00
3 2026-01-01 03:00:00
4 2026-01-01 04:00:00
95 2026-01-04 23:00:00
96 2026-01-05 00:00:00
97 2026-01-05 01:00:00
98 2026-01-05 02:00:00
99 2026-01-05 03:00:00
example of my code
power_data['Day of Week'] = power_data['Date'].dt.dayofweek+1
power_data['Day of Weekv2'] = power_data['Date'].dt.isoweekday
Above is a example portion of my dataframe, the data formatting I would like to follow is Sunday be 1, monday be 2…etc and Saturday be equal to 7. Please let me know if I can do this how its currently presented
As per pandas documentation for weekday, it mentions that:
The day of the week with Monday=0, Sunday=6.
Which means, you just need to add 2 to the weekday value as you want to shift Monday from 0 to 2, then take modulo by 8
# df is your dataframe, and date is the column name consisting pandas Timestamp
>>> (df['date'].dt.weekday+2)%8
#output:
0 5
1 5
2 5
3 5
4 5
5 5
6 5
7 5
8 5
9 5
10 5
11 5
12 5
13 5
14 5
15 5
16 5
17 5
18 5
Name: date, dtype: int64