使用另一个 pandas 数据框的列填充 na 值，但使用列索引，而不是名称

Question

I have a peculiar situation in which I know that the 0th column may contain nan , and in all these cases, the 1st column contains the missing value. I do not know the names of these columns ahead of time, so I want to select them by index instead.

I am able to select the column like so:

df.iloc[:, [0]]

And normally I am able to fill na values from another column like this, if I knew the names:

df["col0"].fillna(df["col1"])

So I thought that I should be able to replicate this by doing this:

df.iloc[:, [0]].fillna(df[:, [1]])

But I get:

Traceback (most recent call last):
  File "/home/---------/.pycharm_helpers/pydev/_pydevd_bundle/pydevd_exec2.py", line 3, in Exec
    exec(exp, global_vars, local_vars)
  File "<input>", line 1, in <module>
  File "/home/---------/_code/microgrid-support/venv/lib/python3.8/site-packages/pandas/core/frame.py", line 3458, in __getitem__
    indexer = self.columns.get_loc(key)
  File "/home/---------/_code/microgrid-support/venv/lib/python3.8/site-packages/pandas/core/indexes/base.py", line 3361, in get_loc
    return self._engine.get_loc(casted_key)
  File "pandas/_libs/index.pyx", line 76, in pandas._libs.index.IndexEngine.get_loc
  File "pandas/_libs/index.pyx", line 82, in pandas._libs.index.IndexEngine.get_loc
TypeError: '(slice(None, None, None), [1])' is an invalid key

How can I fill the na values using the column index instead of its name?

Answer 1

You could use bfill with iloc

import pandas as pd
import numpy as np
df = pd.DataFrame({'a':[np.nan,2,3],'b':[100,200,300],'c':['x','y','z']})

df.iloc[:,:2] = df.iloc[:,:2].bfill(axis=1)

print(df)

Output

     a      b  c
0  100.0  100.0  x
1    2.0  200.0  y
2    3.0  300.0  z

Answer 2

Use the iloc accessor. Slice using index range to avaoid running into issues.

Sample

df1 = pd.DataFrame({'sub_name': [np.nan,'AAB','AAC','BAA','CAA','CAC','CAD','CAE','EAA', 'FAA'], 
'val_1': [2,4,8,7,4,6,2,3,8,3], 
'A':[208,208,208,210,213,213,213,213,222,223]})


df1.iloc[0:1,0].fillna(df1.iloc[0,1])




   sub_name  val_1    A
0        2      2  208
1      AAB      4  208
2      AAC      8  208
3      BAA      7  210
4      CAA      4  213
5      CAC      6  213
6      CAD      2  213
7      CAE      3  213
8      EAA      8  222
9      FAA      3  223

Answer 3

I found a way of doing it myself differently than the other answers. I just accessed the name of the column via the .columns attribute of the dataframe and used that:

For example:

df[df.columns[0]].fillna(df[df.columns[1]])