Group a column when none of the rows are unique into one using pandas

Question

Name Class Marks1 Marks2
AA    CC    10     
AA    CC           33
AA    CC    21     
AA    CC           24

I want to transform data in the above format into

Name Class Marks1  Marks2
AA   CC    10      33
AA   CC    21      24

How should i achieve the result? PS- This is just an example of the data the data is too big and has many more columns and it can have many such rows. How should i remove the duplicate rows without affecting the data quality.

Set the index to the Class column first, then use groupby – Nordle Jul 25 '18 at 12:04 — Nordle, Jul 25 '18 at 12:04

score 0 · Answer 1 · answered Jul 25 '18 at 12:04

0

You could do

In [421]: df.groupby(df.index // 2).first()
Out[421]:
  Name Class  Marks1  Marks2
0   AA    CC    10.0    33.0
1   AA    CC    21.0    24.0

answered Jul 25 '18 at 12:04

Zero

74,117
18
147
154

Might want to use `np.arange` (or similar) just in case `df.index` isn't compatible with this approach. – Jon Clements Jul 25 '18 at 12:23

Group a column when none of the rows are unique into one using pandas

1 Answers1