my expectation that
cheer_leading.nlargest(10, s_cols, keep='all')
will return top performers for each column, but in reality, it returns only performers in the first columns. So to get the largest in every category I have to iterate through columns and save indexes manually.
top_ = set()
for c in s_cols:
top_ |= set(cheer_leading[c].nlargest(10).index)
cheer_leading.loc[top_]
is there an easier way to achieve the same?