python 3.x - Pandas의 다른 탭에서 일치에 따라 값 변경

df1 = pd.read_csv(io.StringIO('''COL1 COL2
A 1
A 2
B 3
B 4
C 5
D 6
E 7
F 8 '''),sep='\s+')
df2 = pd.read_csv(io.StringIO('''Old New 
A  A.1
C  C.1
E  E.1 '''),sep='\s+')
df1['COL1'] = df1.merge(df2.rename(columns={'Old':'COL1'}),on='COL1',how='left')['New'].fillna(df1['COL1'])

산출:

   COL1    COL2
0   A.1 1
1   A.1 2
2   B   3
3   B   4
4   C.1 5
5   D   6
6   E.1 7
7   F   8

사용하다 df.to_dict 와 Series.map :

In [516]: x = df2.set_index('Old').to_dict()['New']
In [519]: df1['COL1'] = df1['COL1'].map(x).fillna(df1['COL1'])
In [521]: df1
Out[521]: 
  COL1  COL2
0  A.1     1
1  A.1     2
2    B     3
3    B     4
4  C.1     5
5    D     6
6  E.1     7
7    F     8

시간 비교:

@adir bargirl의 답변 :

In [540]: def f1():
     ...:     df1['COL1'] = df1.merge(df2.rename(columns={'Old':'COL1'}),on='COL1',how='left')['New'].fillna(df1['COL1'])
     ...: 
In [542]: %timeit f1()
2.09 ms ± 74.2 µs per loop (mean ± std. dev. of 7 runs, 100 loops each)

내 대답 :

In [541]: def f2():
     ...:     x = df2.set_index('Old').to_dict()['New']
     ...:     df1['COL1'] = df1['COL1'].map(x).fillna(df1['COL1'])
     ...: 
In [543]: %timeit f2()
1.06 ms ± 14.8 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)