CODEWITHSUNDEEP
pythonpandas
Jason
Jason

Reputation: 467

Move the value in the second duplicate to the first duplicate

This post is to assign value of last row to first row: Move last value to first value.

I would like to move the value in the second duplicate to the first duplicate and set others to NaT.

ID  OutBedTime        DateOutBed
1   16/05/2018 0:17   16/05/2018
1   16/05/2018 4:05   16/05/2018
1   16/05/2018 6:05   16/05/2018
1   17/05/2018 1:27   17/05/2018
1   17/05/2018 4:41   17/05/2018
1   17/05/2018 5:32   17/05/2018

Expected output

ID  OutBedTime        DateOutBed    TimeOutBedFinal
1   16/05/2018 0:17   16/05/2018    16/05/2018 4:05
1   16/05/2018 4:05   16/05/2018    NaT
1   16/05/2018 6:05   16/05/2018    NaT
1   17/05/2018 1:27   17/05/2018    17/05/2018 4:41
1   17/05/2018 4:41   17/05/2018    NaT
1   17/05/2018 5:32   17/05/2018    NaT

Thank you.

Upvotes: 3

Views: 116

Answers (1)

BENY
BENY

Reputation: 323326

Let us do reindex with apply and select the second of row , then do the same as we did from pervious question

df['New']=df.groupby('DateOutBed')['OutBedTime'].apply(lambda x : x.iloc[[1]]).reset_index(level=1,drop=True).reindex(df.DateOutBed).values
df['New']=df.New.mask(df.DateOutBed.duplicated())
df
   ID      OutBedTime  DateOutBed             New
0   1  16/05/20180:17  16/05/2018  16/05/20184:05
1   1  16/05/20184:05  16/05/2018             NaN
2   1  16/05/20186:05  16/05/2018             NaN
3   1  17/05/20181:27  17/05/2018  17/05/20184:41
4   1  17/05/20184:41  17/05/2018             NaN
5   1  17/05/20185:32  17/05/2018             NaN

Check the update 

df['New']=df.groupby('DateOutBed')['OutBedTime'].transform(lambda x : x.iloc[1] if len(x)>1 else x.iloc[0])
df['New']=df.New.mask(df.DateOutBed.duplicated())

Upvotes: 2

Related Questions