Pandas read_csv: AttributeError: 'NoneType' object has no attribute 'dtype'

Question

Suppose I have a file test.csv which looks like this:

A,B,C
Hello,Hi,1

I'm attempting to read this into a Pandas dataframe:

cols = ['A','B','C']
col_types = {'A': str, 'B': str, 'C': int}
test = pd.read_csv('test.csv', names=cols, dtype=col_types)

This produces the error

AttributeError: 'NoneType' object has no attribute 'dtype'

Any ideas?

Yep, it's definitely the error I'm getting. I'm using '0.12.0', according to pd.__version__. — nilkn
– nilkn, Commented May 21, 2015 at 20:35

chrisaycock · Accepted Answer · 2015-05-21 20:16:49Z

3

Your file already had the header row, so no need to specify any names.

In [6]: test = pd.read_csv('test.csv', dtype=col_types)

In [7]: test
Out[7]:
       A   B  C
0  Hello  Hi  1

In [8]: test.dtypes
Out[8]:
A    object
B    object
C     int64
dtype: object

answered May 21, 2015 at 20:16

chrisaycock

38.1k15 gold badges94 silver badges128 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

Padraic Cunningham Over a year ago

The OP's code should not give the error suggested though no?

chrisaycock Over a year ago

@PadraicCunningham The error I get in pandas 0.15.2 is cannot safely convert passed user dtype of <i8 for object dtyped data in column 2. I just figured the OP was using a different version.

Padraic Cunningham Over a year ago

@chriscock, yes I tried there and got the same, your approach is obviously the correct one, the error just seems a bit strange.

nilkn Over a year ago

Thanks, this seems to fix that error. I'd vote up your answer but I can't as I have < 15 reputation.

Collectives™ on Stack Overflow

Pandas read_csv: AttributeError: 'NoneType' object has no attribute 'dtype'

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related