I was looking to replace all np.nan values in a dataframe with None, I was trying to do this using fillna, but it seems like this is not supported (through fillna, though you can use where):

In [1]: import pandas as pd
i
In [2]: import numpy as np

In [3]: df = pd.DataFrame([[np.nan, 1], [1, np.nan]])

In [4]: df.fillna(value=None)
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-4-f459dbd09698> in <module>()
----> 1 df.fillna(value=None)

/Users/yoshikivazquezbaeza/.virtualenvs/ipy/lib/python2.7/site-packages/pandas/core/frame.pyc in fillna(self, value, method, axis, inplace, limit, downcast, **kwargs)
   2530                                              axis=axis, inplace=inplace,
   2531                                              limit=limit, downcast=downcast,
-> 2532                                              **kwargs)
   2533 
   2534     @Appender(_shared_docs['shift'] % _shared_doc_kwargs)

/Users/yoshikivazquezbaeza/.virtualenvs/ipy/lib/python2.7/site-packages/pandas/core/generic.pyc in fillna(self, value, method, axis, inplace, limit, downcast)
   2522         if value is None:
   2523             if method is None:
-> 2524                 raise ValueError('must specify a fill method or value')
   2525             if self._is_mixed_type and axis == 1:
   2526                 if inplace:

ValueError: must specify a fill method or value

This is what I really was expecting to see:

In [6]: df.where((pd.notnull(df)), None)
Out[6]: 
      0     1
0  None     1
1     1  None

Info about my system:

commit: None
python: 2.7.9.final.0
python-bits: 64
OS: Darwin
OS-release: 14.4.0
machine: x86_64
processor: i386
byteorder: little
LC_ALL: en_US.UTF-8
LANG: en_US.UTF-8

pandas: 0.16.2
nose: 1.3.4
Cython: None
numpy: 1.9.2
scipy: 0.14.0
statsmodels: None
IPython: 3.0.0
sphinx: 1.2.2
patsy: None
dateutil: 2.4.2
pytz: 2015.4
bottleneck: None
tables: None
numexpr: None
matplotlib: 1.4.3
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: None
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: None
pymysql: None
psycopg2: None

Comment From: jreback

in a float column, None is de-facto represented by np.nan (and in most types). So this doesn't make any sense. see the docs here

None makes things into object dtype which is rarely a good idea in any event.

Comment From: ElDeveloper

I see, thanks for the explanation!

On (Aug-20-15|15:19), Jeff Reback wrote:

in a float column, None is de-facto represented by np.nan (and in most types). So this doesn't make any sense. see the docs here

None makes things into object dtype which is rarely a good idea in any event.

Reply to this email directly or view it on GitHub: https://github.com/pydata/pandas/issues/10871#issuecomment-133196500