0

Am trying to get the occurrences of certain pattern based on Index in Dataframe , any help would be appreciated

Have manually added a column as Index and need to get occurrences of pattern based on Index and the column



dataset sample      
a(index)    d   
pattern 1   test    
pattern 1   test    
pattern 1   test2   
pattern 2   test3   
pattern 2   test    
pattern 2   test    

expected output     

Am looking to make a dataframe something like below with the above sample data

pattern  test test2 test3
----------------------------------
pattern 1   2   1        0
pattern 2   2   0        1
Praveen V
  • 37
  • 1
  • 7
  • 2
    `df.groupby(['a','d']).count().reset_index(name='Count')` – anky Aug 13 '19 at 04:47
  • 1
    If a is index, try df.groupby([df.index, 'd']).count().reset_index() – Vaishali Aug 13 '19 at 04:48
  • **Thanks for the quick update , have re-phased the question.. seems i did not express correctly. Any suggestion here would be helpful. Am trying to understand % of certain pre-defined value based on pattern** – Praveen V Aug 13 '19 at 07:00

1 Answers1

0

If you reset the index, it's a pretty straightforward groupby:

In [18]: df
Out[18]:
               d
a
pattern 1   test
pattern 1   test
pattern 1  test2
pattern 2  test3
pattern 2   test
pattern 2   test

In [19]: df.reset_index().groupby(['a', 'd']).apply(len).reset_index()
Out[19]:
           a      d  0
0  pattern 1   test  2
1  pattern 1  test2  1
2  pattern 2   test  2
3  pattern 2  test3  1
Randy
  • 14,349
  • 2
  • 36
  • 42