CODEWITHSUNDEEP

pythonpandasdatetime

Reputation: 113

Combining pandas with datetime

I have a dataframe with start and end dates. I am trying to create a third column with the following conditions:

if dt < 24 hours; return the actual difference between start and end date
if dt > 24 hours; return start date + 24 hours

I have been able to create a column with a 24 hour difference, but I am not able to create a loc-statement that can do the above. Any help?

df2['end_shutdown_analysis'] = df2['Shutdown timestamp'] + timedelta(hours=24)

Upvotes: 0

Views: 58

Answers (1)

Anurag Dabas

Reputation: 24314

you can try via np.where():

import numpy as np

df2['end_shutdown_analysis'] =np.where(
            df2['Shutdown timestamp'].dt.hour<24, # condition
            df['start']-df['end'],                # value if true
            df2['Start']+pd.DateOffset(hours=24)  # else value.
          )

OR

via loc:

m=df2['Shutdown timestamp'].dt.hour<24
df.loc[m,'end_shutdown_analysis']=df['start']-df['end']
df.loc[~m,'end_shutdown_analysis']=df2['start']+pd.DateOffset(hours=24)

Note: you can also use pd.Timedelta(hours=24) in place of pd.DateOffset(hours=24)

Upvotes: 2

Related Questions