On Wed, May 13, 2015 at 3:27 PM, Nathaniel Smith wrote:
> I don't think pandas allows blank values in integer columns? You might get
> better results asking on the pandas list, though -- see
> http://pandas.pydata.org/community.html
>
"integer columns" seems to be the key. they have to be floa
I don't think pandas allows blank values in integer columns? You might get
better results asking on the pandas list, though -- see
http://pandas.pydata.org/community.html
-n
On May 13, 2015 2:17 PM, "Vincent Davis" wrote:
> I have a large (~400mb) csv file I am trying to open in Pandas. When
I have a large (~400mb) csv file I am trying to open in Pandas. When I
don't specify the dtype and open it with the following command It appears
to work.
df = pd.io.parsers.read_csv(CSVFILECLEAN2013, quotechar='"',
low_memory=False, na_values='')
If I try to specify the dtype for each field I ge
