How do I extract only the hours from timedelta?-CodePudding

How do I extract only the hours from this:

TimedeltaIndex(['0 days 00:00:00', '0 days 00:00:00', '0 days 00:00:00',
            '0 days 00:00:00', '0 days 01:00:00', '0 days 01:00:00',
            '0 days 01:00:00', '0 days 01:00:00', '0 days 02:00:00',
            '0 days 02:00:00', '0 days 02:00:00', ... ],
           dtype='timedelta64[ns]', freq=None)

I want only the hours and minutes stored in a column from my data frame.

CodePudding user response：

You can use datetime components.

import pandas as pd
from pandas import TimedeltaIndex
import datetime

df = pd.DataFrame(TimedeltaIndex(['0 days 00:00:00', '0 days 00:00:00', '0 days 00:00:00',
            '0 days 00:00:00', '0 days 01:00:00', '0 days 01:00:00',
            '0 days 01:00:00', '0 days 01:00:00', '0 days 02:00:00',
            '0 days 02:00:00', '0 days 02:00:00'],
           dtype='timedelta64[ns]', freq=None))

df['hours'] = df[0].dt.components['hours']

Output df:

    0               hours
0   0 days 00:00:00 0
1   0 days 00:00:00 0
2   0 days 00:00:00 0
3   0 days 00:00:00 0
4   0 days 01:00:00 1
5   0 days 01:00:00 1
6   0 days 01:00:00 1
7   0 days 01:00:00 1
8   0 days 02:00:00 2
9   0 days 02:00:00 2
10  0 days 02:00:00 2

CodePudding user response：

Use:

temp = ['0 days 00:00:00', '0 days 00:00:00', '0 days 00:00:00',
            '0 days 00:00:00', '0 days 01:00:00', '0 days 01:00:00',
            '0 days 01:00:00', '0 days 01:00:00', '0 days 02:00:00',
            '0 days 02:00:00', '0 days 02:00:00']

import pandas as pd
df = pd.DataFrame(range(len(temp)), index = pd.to_timedelta(temp))
df.index.components['hours']

Output: