How to do value counts with groupby in a numpy array pandas column

Question

I have a pandas dataframe like this:

import pandas as pd
foo = pd.DataFrame({'s': ['a','b','a','a'], 'versions':[['1','2'],'2',['1','2'],'2']})

I would like to value_counts() he columns versions by group s

I have tried

foo.groupby('s')['versions'].value_counts()

but i get an error TypeError: unhashable type: 'list', any ideas ?

jezrael · Accepted Answer · 2021-03-30 08:22:43Z

1

You can convert lists/arrays to hashable tuples, but because there are also strings need if-else statement for avoid split strings 100 to tuple (1,0,0):

foo['versions'] = foo['versions'].apply(lambda x: tuple(x) if isinstance(x, list) else (x,))
print (foo)
   s versions
0  a   (1, 2)
1  b     (2,)
2  a   (1, 2)
3  a     (2,)

s = foo.groupby('s')['versions'].value_counts()
print (s)
s  versions
a  (1, 2)      2
   (2,)        1
b  (2,)        1
Name: versions, dtype: int64

answered Mar 30, 2021 at 8:22

jezrael

868k103 gold badges1.4k silver badges1.3k bronze badges

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

Collectives™ on Stack Overflow

How to do value counts with groupby in a numpy array pandas column

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related