Dividing 24h into working shifts in Python Pandas

I am dealing with dividing a day into working shifts. Let's have a look at my sample code:

I'd like to divide the time into 3 shifts, 00:00 to 08:00 is Shift1, 08:00 to 16:00 will be Shift2 and till 00:00 will be Shift3.

What I get is true, but I would like to know if there is any elegant and easier way to do it?

In: 
import pandas as pd

data = [['24.12.2020 05:17:32', 6], ['24.12.2020 07:20:06', 5],
        ['24.12.2020 09:33:29', 9],['24.12.2020 14:21:54', 12],
        ['24.12.2020 18:52:01', 11],['24.12.2020 22:05:19', 4]]

df = pd.DataFrame(data, columns = ['DateTime', 'Counter'])

print(df)

the dataframe looks like:

Out:
              DateTime  Counter
0  24.12.2020 05:17:32        6
1  24.12.2020 07:20:06        5
2  24.12.2020 09:33:29        9
3  24.12.2020 14:21:54       12
4  24.12.2020 18:52:01       11
5  24.12.2020 22:05:19        4

In:

df['DateTime'] = pd.to_datetime(df['DateTime'])
df['Hour'] = df['DateTime'].dt.hour
df['Hour'] = pd.to_numeric(df['Hour'],downcast="float")

df['Shift1'] = (df['Hour']<8)
df['Shift2'] = (df['Hour']>=8) & (df['Hour']<16)
df['Shift3'] = (df['Hour']>=16)

df['Shift1'] = df['Shift1'].astype(int)
df['Shift2'] = df['Shift2'].astype(int)
df['Shift3'] = df['Shift3'].astype(int)

df['Shift1'] = df['Shift1'].replace([1,0], ['Shift1',''])
df['Shift2'] = df['Shift2'].replace([1,0], ['Shift2',''])
df['Shift3'] = df['Shift3'].replace([1,0], ['Shift3',''])

df['SHIFTS'] = df['Shift1']+df['Shift2']+df['Shift3']

print(df)

Out:
             DateTime  Counter  Hour  Shift1  Shift2  Shift3  SHIFTS
0 2020-12-24 05:17:32        6   5.0  Shift1                  Shift1
1 2020-12-24 07:20:06        5   7.0  Shift1                  Shift1
2 2020-12-24 09:33:29        9   9.0          Shift2          Shift2
3 2020-12-24 14:21:54       12  14.0          Shift2          Shift2
4 2020-12-24 18:52:01       11  18.0                  Shift3  Shift3
5 2020-12-24 22:05:19        4  22.0                  Shift3  Shift3

Solution 1:

Try to create SHIFTS column before then pivot your dataframe:

df['SHIFTS'] = df['DateTime'].astype('datetime64').dt.hour \
                             .floordiv(8).add(1).astype(str).radd('shift')
df = df.join(df.reset_index().pivot('index', 'SHIFTS', 'SHIFTS')).fillna('')

Output:

>>> df
              DateTime  Counter  SHIFTS  shift1  shift2  shift3
0  24.12.2020 05:17:32        6  shift1  shift1                
1  24.12.2020 07:20:06        5  shift1  shift1                
2  24.12.2020 09:33:29        9  shift2          shift2        
3  24.12.2020 14:21:54       12  shift2          shift2        
4  24.12.2020 18:52:01       11  shift3                  shift3
5  24.12.2020 22:05:19        4  shift3                  shift3