Installing numpy on Windows 7 and python 3.x

Question

I can't install a library, or at least numpy. I have a Windows 7 Home Premium 64 bits and Python 3.3.3 (v3.3.3:c3896275c0f6, Nov 18 2013, 21:18:40) [MSC v.1600 32 bit (Intel)] on win32.
I go to their website to Getting Numpy which sends me to SourceForge where I download the latest (I have already update my python to 3.3) and open and install it. Apparently successfully. But when I open python and type import numpy I get:

Traceback (most recent call last):
  File "<pyshell#1>", line 1, in <module>
    import numpy
  File "C:\Python33\lib\site-packages\numpy\__init__.py", line 168, in <module>
    from . import add_newdocs
  File "C:\Python33\lib\site-packages\numpy\add_newdocs.py", line 13, in <module>
    from numpy.lib import add_newdoc
  File "C:\Python33\lib\site-packages\numpy\lib\__init__.py", line 8, in <module>
    from .type_check import *
  File "C:\Python33\lib\site-packages\numpy\lib\type_check.py", line 11, in <module>
    import numpy.core.numeric as _nx
  File "C:\Python33\lib\site-packages\numpy\core\__init__.py", line 6, in <module>
    from . import multiarray
ImportError: DLL load failed: %1 no es una aplicación Win32 válida.

The last line it means that %1 it is not a valid Win32 program.

Tere is a similar question here where they point the OP to the Windows Installer. So I download the zip file and decompress it. And now?

Somewhere I found that it should be in the Lib folder and then run the setup.py file. I tried again to import numpy but again the same error.

I found here some instructions, but the first step it is already different, my error message it is not the same, is the same as step 6. But then the step 8 is to install the 32-bits system after uninstalling numpyand Python, but there is no 32 version for Python 3.3 (or at least not that I understand). What should I do?

So I reinstall the python 3.3 I already had. And this time when I type from numpy import * or import numpy I don't get an error but when I try to make an array I do:

Traceback (most recent call last):
  File "<pyshell#4>", line 1, in <module>
    type(array([1,6,3,7]))
NameError: name 'array' is not defined

Another way I found is from this source with python 3.3 which leads me to another unofficial page where I download the numpy-MKL-1.8.0.win32-py3.3 file following the steps of the blog. But still the same error of NameError.

I didn't restart the computer yet... I forgot something else or I miss one step?

So just to clarify - if you write from numpy import * followed by x = array([1,2,3]), you get a NameError, and no other errors? — senshin
– senshin, Commented Jan 21, 2014 at 18:50
True, but now after rebooting it works fine, I will post it as an answer what I consider it worked for me — llrs
– llrs, Commented Jan 21, 2014 at 18:55

llrs · Accepted Answer · 2014-01-21 19:02:05Z

2

So, after rebooting it seems to work well. I post here what worked for me

Install python for just for me. Download from here the library, install the latest (or the version you want) with 32 bites even if the computer is 64 bites and restart the computer.

answered Jan 21, 2014 at 19:02

llrs

3,40541 silver badges72 bronze badges

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

Comments

John Greenall · Accepted Answer · 2014-01-21 18:57:47Z

0

Christophe Gohlke maitains an extensive collection of windows binaries for various python libraries including numpy. Have you tried here?

/www.lfd.uci.edu/~gohlke/pythonlibs/

answered Jan 21, 2014 at 18:57

John Greenall

1,69013 silver badges17 bronze badges

2 Comments

llrs Over a year ago

This is the final link I posted it worked well but just after restarting the computer

Ben Reynwar Over a year ago

Using the binaries in the link it is not necessary to use the 32 bit version of python (they have binaries for both versions).

Collectives™ on Stack Overflow

Installing numpy on Windows 7 and python 3.x

2 Answers 2

Comments

2 Comments

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

Comments

2 Comments

Linked

Related