What is the correct way of combining two data frames, where one contains duplicates of the other?

Asked Apr 03 '19 at 15:05

Active Apr 03 '19 at 15:05

Viewed 17 times

I'm trying to combine two data frames, where the one contains duplicates of the other data frame. I want the duplicate values to get the same values as the unique ones from the other data frame.

I have tried to use the 'merge' and 'join' function in Pandas, but either give me the desired result.

df1 = pd.DataFrame({'Number': np.arange(3),
                    'Class':['a','b','c']})
df2 = pd.DataFrame({'Number':[0,1,2,2,1,1,0,2,0,0]})

I have two data frames and want to obtain a data frame which extends the class label of df1 on to the duplicate values in df2, so something like:

df3:  {'Number'}  {'Class'}
  0       0            a
  1       1            b
  2       2            c
  3       2            c
  4       1            b
  5       1            b
  6       0            a
  7       2            c
  8       0            a
  9       0            a

asked Apr 03 '19 at 15:05

Kristian Nielsen

Check merge :-) – BENY Apr 03 '19 at 15:08
@Wen-Ben As stated at the beginning of my question, I have tried 'merge', and it does not give me the answer I expected. So, would you mind elaborating a bit? I'll be happy to update my question with the specifications of the merge function that I tried. – Kristian Nielsen Apr 03 '19 at 15:16
1

did you try `df2.merge(df1,how='left') ` – BENY Apr 03 '19 at 15:16
No I did not, but it does the trick! I thought it had to be a lot more complicated. – Kristian Nielsen Apr 03 '19 at 15:20
I did pd.merge(df1,df2,on = ['Number''],how='inner') – Kristian Nielsen Apr 03 '19 at 15:21
1

Mission complete – BENY Apr 03 '19 at 15:22
@Wen-Ben The resulting data frame is longer than the longest of the two, do you have an idea why this is the case? My application is identical to the above example, both with a data frame of much longer length. – Kristian Nielsen Apr 03 '19 at 17:04
Both of them have duplicate row, so they do a cross join – BENY Apr 03 '19 at 17:05
Thank you!! The second data frame was not meant to have duplicates but no I see that it indeed does! – Kristian Nielsen Apr 03 '19 at 17:19

What is the correct way of combining two data frames, where one contains duplicates of the other?

0 Answers0