check if DataFrame column is boolean type

Question

I have a pandas DataFrame as below:

>>> df
t                        v
2014-02-21 10:30:43  False
2014-02-21 10:31:34  False
2014-02-21 10:32:25  False

>>> df.dtypes
t   str
v   bool

I need to check if the dtype of this dataframe is bool. I tried with:

>>> print('Type is Boolean = ', isinstance(df1, bool) )
Type is Boolean = False

How can I do this?

Source: check if variable is dataframe

Reference `dtype` directly. – Jakob Bowyer Feb 24 '14 at 08:55 — Jakob Bowyer, Feb 24 '14 at 08:55
How? Any sample code please? – curlyreggie Feb 24 '14 at 08:59 — curlyreggie, Feb 24 '14 at 08:59

score 8 · Accepted Answer · edited Mar 18 '23 at 14:02

8

You can print the dtypes of the columns:

In [2]:

import pandas as pd
 
df = pd.DataFrame({'a':[True,False,False]})
df
Out[2]:
       a
0   True
1  False
2  False

[3 rows x 1 columns]

In [3]:

df.dtypes

Out[3]:
a    bool
dtype: object
In [4]:

df.a.dtypes
Out[4]:
dtype('bool')

In your case, df1.v.dtypes should print the same output as above

The other thing to note that isinstance(df, bool) will not work as it is a pandas dataframe or more accurately:

In [7]:

type(df)
Out[7]:
pandas.core.frame.DataFrame

The important thing to note is that dtypes is in fact a numpy.dtype you can do this to compare the name of the type with a string but I think isinstance is clearer and preferable in my opinion:

In [13]:

df.a.dtypes.name == 'bool'
Out[13]:
True

edited Mar 18 '23 at 14:02

double-beep

5,031
17
33
41

answered Feb 24 '14 at 09:13

EdChum

376,765
198
813
562

That's what I did. I need to check if `df.a.dtypes == bool`. How can I do this? – curlyreggie Feb 24 '14 at 09:57
1

Okay. I had to do this: `if df1.v.dtype == 'bool':` to check if it is a boolean DataFrame. Thanks! – curlyreggie Feb 24 '14 at 10:04
2

@curlyreggie `dytpes` is in fact a `numpy.dtype` see: http://docs.scipy.org/doc/numpy/reference/generated/numpy.dtype.html, you could also do `df.a.dtypes.name == 'bool'` would yield `True` – EdChum Feb 24 '14 at 10:12
Can we do something like `isinstance(df[col].dtype, bool)`? – information_interchange Oct 29 '21 at 21:16

score 0 · Answer 2 · answered Mar 17 '23 at 06:11

To get the dtype of a specific column, you have two ways:

Use DataFrame.dtypes which returns a Series whose index is the column header.

$ df.dtypes.loc['v']

bool

Use Series.dtype or Series.dtypes to get the dtype of a column. Internally Series.dtypes calls Series.dtype to get the result, so they are the same.

$ df['v'].dtype

bool

$ df['v'].dtypes

bool

All of the results return the same type

$ type(df.dtypes.loc['v'])

<class 'numpy.dtype[bool_]'>

$ type(df['v'].dtype)

<class 'numpy.dtype[bool_]'>

To check if it is a bool type also has multiple ways

$ df['v'].dtype == 'bool'

True

$ np.issubdtype(df['v'].dtype, bool)

True

$ df['v'].dtype.type is np.bool_

True

You can also select the columns with specific types with DataFrame.select_dtypes

$ df.select_dtypes('bool')

       v
0  False
1  False
2  False

check if DataFrame column is boolean type

2 Answers2