Question

问题：我有2个数据框df1和df2。我的目标是通过替换df1中的部分值来修改df2。

import pandas as pd

# dataframe 1
data = {'A':[90,20,30,25,50,60],
        'B':['qq','ee','rr','tt','ii','oo'],
        'C':['XX','VV','BB','NN','KK','JJ']}
df1 = pd.DataFrame(data)

# dataframe 2
convert_table = {'X': ['dd','ee','ff','gg','hh','ii','ll','mm','nn','oo','pp','qq','rr','ss','tt','uu'], 
                 'Y': ['DD','VV','FF','GG','HH','KK','LL','MM','NN','JJ','PP','XX','BB','SS','NN','LL'], 
                 'Z': [5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61]}
df2 = pd.DataFrame(convert_table)

# search values of df1 inside of df2 and replace values
for idx1,row1 in df1.iterrows():
    for idx2, row2 in df2.iterrows():
        if row1['B']==row2['X'] and row1['C']==row2['Y']:
            df1.replace(to_replace=row1['B'],value=row2['Z'],inplace=True)

正如您所看到的，我有2个for循环，并检查df1（row1的通用行是否在df2内找到。如果满足此条件，则我将row1 ['B']中包含的值替换为row2['Z']

中包含的值

因此，我得到的结果（正是我希望得到的结果）：

In [120]: df1
Out[120]: 
    A   B   C
0  90  43  XX
1  20   7  VV
2  30  47  BB
3  25  59  NN
4  50  19  KK
5  60  37  JJ

注意B列是如何变化的。

问题：您能否建议我更好地编写代码？我想尽可能快地使用Pandas或Python提供的内置函数。

注意：数据框中包含的数据仅用于演示目的。

Answer 1

在两列上使用合并：

df1.merge(df2, left_on=['B','C'], right_on=['X','Y'], how='left')

how='left'在这里至关重要。如果您不明白原因，请阅读Brief primer on merge methods (relational algebra)。

我将修改您的示例以创建一个df1中的条目，该条目在df2中不存在，即('ii','KK')

In [1]:
# dataframe 2
convert_table = {'X': ['dd','ee','ff','gg','hh','ll','mm','nn','oo','pp','qq','rr','ss','tt','uu'], 
                 'Y': ['DD','VV','FF','GG','HH','LL','MM','NN','JJ','PP','XX','BB','SS','NN','LL'], 
                 'Z': [5,7,11,13,17,19,23,29,37,41,43,47,53,59,61]}
df2 = pd.DataFrame(convert_table)



In [2]: merged = df1.merge(df2, left_on=['B','C'], right_on=['X','Y'], how='left')
        merged
Out[2]: 
    A   B   C    X    Y     Z
0  90  qq  XX   qq   XX  43.0
1  20  ee  VV   ee   VV   7.0
2  30  rr  BB   rr   BB  47.0
3  25  tt  NN   tt   NN  59.0
4  50  ii  KK  NaN  NaN   NaN
5  60  oo  JJ   oo   JJ  37.0

现在检索最终的数据帧：

In [3]:
merged.ix[merged.Z.notnull(),'B'] = merged.ix[merged.Z.notnull(),'Z']
merged = merged[['A','B','C']]
merged

Out[3]:
    A   B   C
0  90  43  XX
1  20   7  VV
2  30  47  BB
3  25  59  NN
4  50  ii  KK
5  60  37  JJ

将值映射到Pandas中不同数据帧的数据帧

1 个答案: