熊猫插值（）向后数据帧

Question

展望未来，interpolate的伟大工程：

 name days 
0  a  NaN 
1  a  NaN 
2  a   2 
3  a   3 
4  a  NaN 
5  a  NaN 

records.loc[:, 'days'].interpolate(method='linear', inplace=True) 

     name days 
0  a  NaN 
1  a  NaN 
2  a   2 
3  a   3 
4  a   4 
5  a   5 

...但是，它没有解决的开始行（仅前进）。 limit_direction参数允许{‘forward’, ‘backward’, ‘both’}。这些都不起作用。有没有适当的方法来向后插入？

我们可以假设一系列递增或递减1，它可能不会从0开始，正如在本例中发生的那样。

Answer 1

看来它只能与参数limit看到docs [In 47]：

添加limit_direction关键字参数与限制工作，使插值填写NaN值向前，向后，或两者（GH9218， GH10420，GH11115）

records = pd.DataFrame(
{'name': {0: 'a', 1: 'a', 2: 'a', 3: 'a', 4: 'a', 5: 'a', 6: 'a', 7: 'a', 8: 'a', 9: 'a'}, 
'days': {0: 0.0, 1: np.nan, 2: np.nan, 3: np.nan, 4: 4.0, 5: 5.0, 6: np.nan, 7: np.nan, 8: np.nan, 9: 9.0}}, 
columns=['name','days']) 

print (records) 
    name days 
0 a 0.0 
1 a NaN 
2 a NaN 
3 a NaN 
4 a 4.0 
5 a 5.0 
6 a NaN 
7 a NaN 
8 a NaN 
9 a 9.0 

#by default limit_direction='forward' 
records['forw'] = records['days'].interpolate(method='linear', 
               limit=1) 
records['backw'] = records['days'].interpolate(method='linear', 
               limit_direction='backward', 
               limit=1) 
records['both'] = records['days'].interpolate(method='linear', 
               limit_direction='both', 
               limit=1) 
print (records) 
    name days forw backw both 
0 a 0.0 0.0 0.0 0.0 
1 a NaN 1.0 NaN 1.0 
2 a NaN NaN NaN NaN 
3 a NaN NaN 3.0 3.0 
4 a 4.0 4.0 4.0 4.0 
5 a 5.0 5.0 5.0 5.0 
6 a NaN 6.0 NaN 6.0 
7 a NaN NaN NaN NaN 
8 a NaN NaN 8.0 8.0 
9 a 9.0 9.0 9.0 9.0