python - 如何只保留数据框列中的字符串

Question

我有一个包含许多公司特征的数据框。我的问题是，在“设施名称”列中，我发现有些行具有浮点值，因此我想删除它们。为此，我使用了以下似乎不起作用的代码行:

df = df[isinstance(df['Facility Name'], str)]

知道怎么做吗？

Answer 1

你很接近。这应该有效:

python 2.x:

df = df[df['Facility Name'].apply(lambda x: isinstance(x, basestring))]

Python 3.x:

df = df[df['Facility Name'].apply(lambda x: isinstance(x, (str, bytes)))]

下面是一些 Python 3.x 代码，您可以看到它是如何工作的:

In [1]: import pandas as pd

In [2]: import numpy as np

In [3]: df = pd.DataFrame({'Facility Name': [1, 0, None, 'Yes', 'No', 'Maybe So', b'what', np.nan, np.inf]})                                                         

In [4]: df                                                                                                                                                           
Out[4]: 
  Facility Name
0             1
1             0
2          None
3           Yes
4            No
5      Maybe So
6       b'what'
7           NaN
8           inf

In [5]: df[df['Facility Name'].apply(lambda x: isinstance(x, str))]                                                                                                  
Out[5]: 
  Facility Name
3           Yes
4            No
5      Maybe So

In [6]: df[df['Facility Name'].apply(lambda x: isinstance(x, (str, bytes)))]                                                                                         
Out[6]: 
  Facility Name
3           Yes
4            No
5      Maybe So
6       b'what'