如何利用Pandas，將一個(gè)文件的某列值根據(jù)另一個(gè)文件對(duì)應(yīng)列名進(jìn)行重新賦值？

為提升性能，應(yīng)減少 pandas.DataFrame.apply() 的逐行操作，在本例中可改用 numpy.where() 二元操作符，如下

In [1]: import pandas as pd

In [2]: import numpy as np

In [3]: df1 = pd.DataFrame({'t': [1,2,3], 'user_id': [10,20,30], 'v': [1.1,2.2,3.3]})

In [4]: df1
Out[4]: 
   t  user_id    v
0  1       10  1.1
1  2       20  2.2
2  3       30  3.3

In [5]: df2 = pd.DataFrame({'t': [4,1,2], 'user_id': [40,10,20], 'v': [400,100,200]})

In [6]: df2
Out[6]: 
   t  user_id    v
0  4       40  400
1  1       10  100
2  2       20  200

In [7]: df3 = pd.merge(df1, df2, how='right', on=['t', 'user_id'])

In [8]: df3
Out[8]: 
   t  user_id  v_x  v_y
0  1       10  1.1  100
1  2       20  2.2  200
2  4       40  NaN  400

In [9]: df3['v'] = np.where(np.isnan(df3.v_x), df3.v_y, df3.v_x)

In [10]: df3
Out[10]: 
   t  user_id  v_x  v_y      v
0  1       10  1.1  100    1.1
1  2       20  2.2  200    2.2
2  4       40  NaN  400  400.0

In [11]: del df3['v_x']

In [12]: del df3['v_y']

In [13]: df3
Out[13]: 
   t  user_id      v
0  1       10    1.1
1  2       20    2.2
2  4       40  400.0

import pandas as pd
import numpy as np
df1 = pd.DataFrame([[0, 1, 2, 3], [4, 5, 6, 7]], columns=['t', 'user_id', 'x', 'y'])
>>> df1
   t  user_id  x  y
0  0        1  2  3
1  4        5  6  7
df2 = pd.DataFrame([[0, 1, 0, 0], [4, 5, 0, 0], [5, 6, 7, 8]], columns=['t', 'user_id', 'x', 'y'])
>>> df2
   t  user_id  x  y
0  0        1  0  0
1  4        5  0  0
2  5        6  7  8
def f(row):
    if np.isnan(row['x_y']):
        row['x'], row['y'] = row['x_x'], row['y_x']
    else:
        row['x'], row['y'] = row['x_y'], row['y_y']
    return row

res = pd.merge(df2, df1, how='left', on=['t', 'user_id']).apply(f, axis=1)[['t', 'user_id', 'x', 'y']]
>>> res
     t  user_id    x    y
0  0.0      1.0  2.0  3.0
1  4.0      5.0  6.0  7.0
2  5.0      6.0  7.0  8.0