Python pandas sort groupby results

Question

I'm successfully extracting and counting data from my dataframe but I'd like to sort on the resulting count values.

groupby snippet:

my_df.groupby('name').Id.nunique().head(5)

Results:

Name
nameA        4
nameB        1
nameC        1
nameD        6
nameE        2

But I'd like to order the results as follows;

Name
nameD        6
nameA        4
nameE        2
nameB        1
nameC        1

I'm assuming this should be straight forward and I'm hoping someone can point me in the right direction.

Honestly you can just use google and probably find your answer. — Tadhg McDonald-Jensen
– Tadhg McDonald-Jensen, Commented Apr 13, 2016 at 22:07

Alexander · Accepted Answer · 2016-04-13 22:10:57Z

2

You don't even need the groupby. Just use value_counts().

>>> df.Name.value_counts()
nameD    6
nameA    4
nameE    2
nameC    1
nameB    1
Name: Name, dtype: int64

answered Apr 13, 2016 at 22:05

Alexander

111k32 gold badges212 silver badges208 bronze badges

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

2 Comments

This should be marked as the answer if there is guaranteed to be 1 unique id per name.

That's great! Thank you!

ej_f · Accepted Answer · 2016-04-13 22:17:08Z

1

Try with:

my_df.groupby('name').Id.nunique().sort_values(ascending=False)

answered Apr 13, 2016 at 22:17

ej_f

4603 silver badges9 bronze badges

SpeedCoder5 · Accepted Answer · 2016-04-13 22:27:05Z

0

@ej_f - right on. Results for me were funny with nunique so I used count instead.

my_df.groupby('name').Id.count().sort_values(ascending=False)

answered Apr 13, 2016 at 22:27

SpeedCoder5

9,1987 gold badges36 silver badges35 bronze badges