使用Pandas在Python中产生有趣的数据帧

Question

I have two DataFrames: A and B. A comtains two columns: Number and Letter, B also contains two columns: Number1 and Number2, and Number contains same data as in both Number1 and Number2. How can I make a ned dataframe which will have column which contains Number1, Number2 and column with its Letter?

 A
Number Letter
1        e
2        l
3        o
4        s
5        p
6        w 
7        r
8        i
9        u

B
Number1 Number2
1           7
4           9
2           1

Needed output (without dublicats):

Number Letter
    1     e
    4     s
    2     l
    7     r
    9     u

If there was only one column in B, i would make it with "merge", but how to be here? Can anybody help?

Answer 1

you can use stack() in conjunction with isin() :

In [109]: A[A['Number'].isin(B.stack())]
Out[109]:
   Number Letter
0       1      e
1       2      l
3       4      s
6       7      r
8       9      u