Ignore warning Pandas KeyError: value not in index

user_11610998 0
Viewed 2756

Is there a way to suppress the pandas KeyError: '[x]' not in index? For example, if I have a data frame with columns A B C, and I call df[['A','B','C','D']], is it possible to have it just return A,B,C and ignore D if it does not exist?

Example code

import pandas as pd
import numpy as np

a = np.matrix('[1,4,5];[1,2,2];[9,7,5]')

df = pd.DataFrame(a,columns=['A','B','C'])

df[['A','B','C','D']]

Here's the error message

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/lib/python3/dist-packages/pandas/core/frame.py", line 2133, in __getitem__
    return self._getitem_array(key)
  File "/usr/lib/python3/dist-packages/pandas/core/frame.py", line 2177, in _getitem_array
    indexer = self.loc._convert_to_indexer(key, axis=1)
  File "/usr/lib/python3/dist-packages/pandas/core/indexing.py", line 1269, in _convert_to_indexer
    .format(mask=objarr[mask]))
KeyError: "['D'] not in index"
pandas warnings
2 Answers

Use the column intersection with your desired list when selecting columns. You get all columns when they exist and only the subset that exists with fewer columns, without any errors. 

l = ['A', 'B', 'C', 'D']
df[df.columns.intersection(l)]

   A  B  C
0  1  4  5
1  1  2  2
2  9  7  5
user_43333590

If you definitely want D, you can reindex() on axis=1:

l = ['A', 'B', 'C', 'D']
df.reindex(l, axis=1)

   A  B  C   D
0  1  4  5 NaN
1  1  2  2 NaN
2  9  7  5 NaN
user_98406370
Related