DataFrame如何根据一列来计算另一列出现的次数

df = pd.DataFrame({'a':[1,1,1,3,3,3,3,3,3,3,3,8],'u':[99,98,67,65,63,67,57,55,51,53,53,55]})

df
Out[72]:
a   u
0   1  99
1   1  98
2   1  67
3   3  65
4   3  63
5   3  67
6   3  57
7   3  55
8   3  51
9   3  53
10  3  53
11  8  55

gropus=df.groupby(['a'])
row = {'a':[],'u':[],'第几次出现':[]}
for k,group in gropus:
row['a'] += group.a.values.tolist()
row['u'] += group.u.values.tolist()
row['第几次申请'] += ((group.index-group.index[0])%group.index.size+1).values.tolist()

pd.DataFrame(row)

Out[73]:
a   u  第几次出现
0   1  99      1
1   1  98      2
2   1  67      3
3   3  65      1
4   3  63      2
5   3  67      3
6   3  57      4
7   3  55      5
8   3  51      6
9   3  53      7
10  3  53      8
11  8  55      1