How to calculate the SUM of time in hh:mi format in pandas dataframe-CodePudding

I have a pandas data frame like this:

how can i calculate the SUM of total time per group? the time format is in HH:MI. for example for G1 we've total (01:30 02:30 05:30)=09:30. I've checked many solutions like separating each item via ":" by for loop or multiplying it *60 but no success.

CodePudding user response：

You can convert each value to Timedelta than sum them:

df['SUM'] = pd.to_timedelta(df['HOURS'].str.split(',').explode()   [':00']) \
              .groupby(level=0).sum()
print(df)

# Output
  GROUP              HOURS             SUM
0    G1  01:30,02:30,05:30 0 days 09:30:00
1    G2  02:30,05:30,07:30 0 days 15:30:00