ValueError: The elements are 0-sized

arrays, numpy, python, python-2.7, text-files

Solution

You're probably better off using `np.genfromtxt` for text files; `np.fromfile` is better for binary files.

This gives the string array you originally seemed to be going for:

>>> np.genfromtxt('tmp.txt', dtype=str)
array([['1', 'Hydrogen', '1.008'],
       ['2', 'Helium', '4.002602'],
       ['3', 'Lithium', '6.94'],
       ['4', 'Beryllium', '9.0121831'],
       ['5', 'Boron', '10.81'],
       ['6', 'Carbon', '12.011']], 
      dtype='|S9')

And this gives you a record array that has flexible dtype:

>>> np.genfromtxt('tmp.txt', dtype=None)
array([(1, 'Hydrogen', 1.008), (2, 'Helium', 4.002602),
       (3, 'Lithium', 6.94), (4, 'Beryllium', 9.0121831),
       (5, 'Boron', 10.81), (6, 'Carbon', 12.011)], 
      dtype=[('f0', '<i8'), ('f1', 'S9'), ('f2', '<f8')])

But this is what I would do, using `unpack` allows you to assign to several variables:

>>> n, name, mass = np.genfromtxt('tmp.txt', dtype='S', unpack=True)
>>> n = n.astype(int)
>>> mass = mass.astype(float)

Problem

Haven't been able to find any information on the web about this problem. I'm trying to read from a text file using numpy. ``` data = np.fromfile("C:\Users\Dan\Desktop\elements.txt",dtype=str,count=-1,sep=' ') ``` When I run this code I get this error: ``` ValueError: The elements are 0-sized. ``` I've never seen this error before, and a Google search didn't return any information about this error. Why is this error occurring, and how can I get around it? EDIT: Here is an excerpt from the text file ``` 1 Hydrogen 1.008 2 Helium 4.002602 3 Lithium 6.94 4 Beryllium 9.0121831 5 Boron 10.81 6 Carbon 12.011 ```

Original source