[SciPy-user] Installing the SVM sandbox

Nils Wagner nwagner at iam.uni-stuttgart.de
Fri Apr 20 04:56:02 EDT 2007


Jose Gomez wrote:
> Hi,
> I am interested in testing out the SVM bit in the sandbox. However, I don't seem
> to be able to understand how to do it. I have installed the scipy package in
> Kubuntu (6.10). I have downloaded the source tarball, and went into the relevant
> directory, compiled the libsvm C++ code, copied the .so and .os files to the
> main sandbox/svm directory, issued a python setup.py build and python setup.py
> install, and tried the tests in the tests directory. They work. However, if I
> try to import svm from my IPython shell, I get this:
>
> |from svm.dataset import LibSvmRegressionDataSet, LibSvmTestDataSet
> ---------------------------------------------------------------------------
> exceptions.ImportError                               Traceback (most recent call
> last)
>
> /home/cuenta/<ipython console>
>
> ImportError: No module named svm.dataset
>
> I must have forgotten something :)
>
>
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.org
> http://projects.scipy.org/mailman/listinfo/scipy-user
>   
Hi,

I am not familiar with Kubuntu but I was able to install the sandbox
package svm
using scipy/Lib/sandbox/enabled_packages.txt on SuSE Linux 10.x

Just add a line with

svm

to that file and reinstall it. (rm -rf build; python setup.py install)

Then you can do something like

>>> from scipy.sandbox import svm
>>>       


HTH,
              Nils


 




More information about the SciPy-User mailing list