How to keep top [2-31] rows of each group in a pandas data frame?

Question

This anwser said that I can use df.groupby('id')['value'].nlargest(30) to get top30 rows for each group. But how to get the rows form top2 to top31 for each group? Is there a function that can do the similar thing like pandas.Series.nlargest

Answer 1

尝试apply .iloc和.iloc来获取，例如您想要1:30

df.sort_values('value').groupby('id').value.apply(lambda x : x.iloc[1:30])