将一个数据帧列的值替换为基于条件的另一个数据帧列的值

Question

import pandas as pd
import numpy as np

dt=pd.DataFrame({'schoolNum':['12','13','14','13'],'oldDistrict':['A','B', 'C','B'],'newDistrict':['F',' ','G',' ']})

dt=pd.DataFrame({'schoolNum':['12','13','14','13'],'oldDistrict':['A','B', 'C','B'],'newDistrict':['F','B','G','B']})

我试过：

dt.ix[dt['oldDistrict']=='13', 'newDistrict']=dt.oldDistrict

但我收到一条关于axis的错误信息。你知道吗

Answer 1

两件事：

1. 要使用.loc而不是.ix:

   dt.loc[<condition>, <column(s) returned>]
   

2. “oldDistrict”没有等于“13”的值-这些值在“schoolNum”中。所以你想要的是：

   dt.loc[dt['schoolNum']=='13', 'newDistrict'] = dt['oldDistrict']
   

Answer 2

你可以用np.nan公司然后先用联合收割机

dt = dt.replace({' ': np.nan})
dt['newDistrict'] = dt['newDistrict'].combine_first(dt['oldDistrict'])

退货

    newDistrict oldDistrict schoolNum
0   F           A           12
1   B           B           13
2   G           C           14
3   B           B           13