如何将Pandas DataFrame groupby结果转换为DataFrame？

Question

Here is what I have done:

df = pd.DataFrame(data={'in':[2, 2, 3, 4], 'out':[2,2,4,4]})
in_out=df.groupby(['in','out']).size()

I got results:

in  out
2   2      2
3   4      1
4   4      1
dtype: int64

How do I turn the results into a dataframe with columns in, out and count?

Answer 1

Use reset_index with parameter name :

df = pd.DataFrame(data={'in':[2, 2, 3, 4], 'out':[2,2,4,4]})
in_out=df.groupby(['in','out']).size().reset_index(name='count')
print in_out
   in  out  count
0   2    2      2
1   3    4      1
2   4    4      1