python - Convert time to categorical variable

Question

df 
Time      | Day_of_the_Week 
16:24:18  | Sat
17:00:01  | Sun
03:48:12  | Mon

Expected Output:

df 
Time      | Day_of_the_Week | Time_Category 
16:24:18  | Sat             |    Afternoon
17:00:01  | Sun             |    Evening
03:48:12  | Mon             |    Midnight

df['Time'][1] returns a "datetime.time"

The following code returns an invalid syntax.

for a in df:
    if df['Time'] > 17:00:00:
         df['Time_Category'] == 'Evening'
    elif df['Time'] > '12:00:00':
         df['Time_Category'] == 'Afternoon'
    elif df['Time'] > '04:00:00':
         df['Time_Category'] == 'Morning'
    else:
        df['Time_Category'] == 'Midnight'

标签： pythonpython-3.xregexpandasdatetime

Answer 1

We can try pd.cut

s = pd.cut(df.Time, pd.to_timedelta(['04:00:00','12:00:00','17:00:00','23:59:59']),
          labels=['Morning','Afternoon','Evening']).astype(str).replace('nan','Midnight')
Out[43]: 
0    Afternoon
1      Evening
2     Midnight
Name: Time, dtype: object
df['Time_category'] = s