Group by multiple columns using multiple keys

Question

This is the structure of my dataframe-

Key1 Key2 Value1 Value2
A    Alpha 16    12345
B    Beta  12    123
A    Alpha 15    1456
A    Beta  14    12345

I have to club Value 1 and Value 2 basis of unique combination of Key 1 and Key 2. I want my final table as follows:

Key1 Key2  Value1    Value2
A    Alpha {16,15}   {12345,1456}
B    Beta  {12}      {123}
A    Beta  {14}      {12345}

Kindly suggest a code. Thanks much appreciated. Coding level- 8 days old.

Ted Petrou · Accepted Answer · 2016-12-16 05:38:32Z

1

You must write your own custom aggregation function. agg gets passed every series that is not a grouping column and returns a single value. Here we use set as the aggregator.

df.groupby(['Key1', 'Key2']).agg(lambda x: set(x.values))

  Key1   Key2    Value1         Value2
0    A  Alpha  {16, 15}  {1456, 12345}
1    A   Beta      {14}        {12345}
2    B   Beta      {12}          {123}

edited Dec 16, 2016 at 5:38

answered Dec 16, 2016 at 5:30

Ted Petrou

62.4k19 gold badges139 silver badges139 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

Piyush Jha Over a year ago

Thanks, this had worked. Couldn't vote (no reputation it seems)

Collectives™ on Stack Overflow

Group by multiple columns using multiple keys

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related