Numpy import from CSV and convert columns to float [duplicate]

Question

ok, so i have a numpy array loaded from a CSV file, the array looks like :

array([['0', '3', '22', ..., '7.25', '1', '0'],
       ['1', '1', '38', ..., '71.2833', '0', '0'],
       ['1', '3', '26', ..., '7.925', '1', '0'],
       ..., 
       ['0', '3', '', ..., '23.45', '1', '0'],
       ['1', '1', '26', ..., '30', '0', '0'],
       ['0', '3', '32', ..., '7.75', '0', '0']], 
      dtype='|S8')

I want to convert the array elements to float but i'm having this error

data2 = np.array(data).astype(np.float)

Traceback (most recent call last):
  File "<input>", line 1, in <module>
ValueError: could not convert string to float:

is there a way to solve this problem with numpy or pandas ?

I think you should look at the root of your problem: how you read in the CSV file. Because this should be able to handle this at once and reading it in as floats. Can you give an extract of how your CSV file looks like? — joris
– joris, Commented May 11, 2013 at 14:46

pradyunsg · Accepted Answer · 2013-05-11 14:55:25Z

3

I think there is an empty string('') in your array. So, change your array's '' into 0s.

Assuming your array is a:

>>> a[a=='']='0'
>>> a2 = a.astype(np.float)

edited May 11, 2013 at 14:55

answered May 11, 2013 at 14:00

pradyunsg

19.9k11 gold badges48 silver badges99 bronze badges

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

Comments

pradyunsg · Accepted Answer · 2013-05-11 14:52:20Z

1

The problem is with a specific value rather than Numpy or your data in general. I think it's the empty string that is causing the problem, there is no way to represent '' as a float.

edited May 11, 2013 at 14:52

pradyunsg

19.9k11 gold badges48 silver badges99 bronze badges

answered May 11, 2013 at 12:58

Tom Dalton

6,19027 silver badges36 bronze badges

Collectives™ on Stack Overflow

Numpy import from CSV and convert columns to float [duplicate]

2 Answers 2

Comments

Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

Comments

Linked

Related