Remove the days in the timedelta object

Question

I have a column in a pandas dataframe that is created after subtracting two times. I now have a timedelta object like this -1 days +02:45:00. I just need to remove the -1 days and want it to be 02:45:00. Is there a way to do this?

Answer 1

I think you can subtract days converted to timedeltas:

td = pd.to_timedelta(['-1 days +02:45:00','1 days +02:45:00','0 days +02:45:00'])
df = pd.DataFrame({'td': td})

df['td'] = df['td'] - pd.to_timedelta(df['td'].dt.days, unit='d')

print (df.head())

        td
0 02:45:00
1 02:45:00
2 02:45:00

print (type(df.loc[0, 'td']))
<class 'pandas._libs.tslibs.timedeltas.Timedelta'>

Or convert timedeltas to strings and extract strings between days and .:

df['td'] = df['td'].astype(str).str.extract('days (.*?)\.')
print (df.head())
          td
0  +02:45:00
1   02:45:00
2   02:45:00

print (type(df.loc[0, 'td']))
<class 'str'>

Answer 2

I found this method easy, others didnt work for me

df['column'] = df['column'].astype(str).map(lambda x: x[7:])

It slices of the days part and you only get time part

Answer 3

If your column is named time1, you can do it like this:

import pandas as pd
import datetime as dt
df['time1'] = pd.to_datetime(str(df.time1)[11:19]) #this slice can be adjusted
df['time1'] = df.time1.dt.time

this is going to convert the timedelta to str, slice the time part from it, convert it to datetime and extract the time from that.

Answer 4

I found a very easy solution for other people who may encounter this problem:

if timedelta_obj.days < 0:
    timedelta_obj.days = datetime.timedelta(
        seconds=timedelta_obj.total_seconds() + 3600*24)

Answer 5

I think what you want to do is to minus two timestamps while ignoring the date. One simple way to do that is to set the two timestamps with the same date before minus.

import pandas as pd    
start_time = pd.Timestamp(2007, 12, 5, 21, 24, 36)
end_time = pd.Timestamp(2007, 12, 6, 20, 24, 36)
start_time - end_time # which returns Timedelta('-1 days +01:00:00')

What you can do is just set end_time with the same date as start_time:

import pandas as pd    
start_time = pd.Timestamp(2007, 12, 5, 21, 24, 36)
end_time = pd.Timestamp(2007, 12, 6, 20, 24, 36)
end_time = end_time.replace(year=start_time.year, month=start_time.month, day=start_time.day)
start_time - end_time # which returns Timedelta('0 days +01:00:00')