On Thursday 20 November 2008 11:11:14 Hans Meine wrote:

I have a 2D matrix comprising a sequence of vectors, and I want to compute the norm of each vector. np.linalg.norm seems to be the best bet, but it does not support axis. Wouldn't this be a nice feature?

Here's a basic implementation. docstring + tests not updated yet, also I wonder whether axis should be the first argument, but that could create compatibility problems.

Ciao, Hans

On Donnerstag 20 November 2008, Alan G Isaac wrote:

On 11/20/2008 5:11 AM Hans Meine apparently wrote:

...

I have a 2D matrix comprising a sequence of vectors, and I want to compute the norm of each vector. np.linalg.norm seems to be the best bet, but it does not support axis. Wouldn't this be a nice feature?

Of possible use until then: http://docs.scipy.org/doc/numpy/reference/generated/numpy.apply_along_axis.h...

Thanks for the hint, yes as you see I have already patched norm() in the meantime.

BTW: Wow, this is an exceptionally nice doc page, sphinx + scipy's doc system really rocks! :-)

Ciao, / / .o. /--/ ..o / / ANS ooo

2009/2/9 Hans Meine meine@informatik.uni-hamburg.de:

...

Here's a basic implementation. docstring + tests not updated yet, also I wonder whether axis should be the first argument, but that could create compatibility problems.

AFAICS, I never received an answer, but IMHO this should be integrated into NumPy. Any objections?

I often work with vectors inside an array, so I would find it useful (even if it is just a two-line wrapper). I'd also add an "out" argument, so that the signature is similar to that of max, min, etc.

norm(a, ord=None, axis=None, out=None)

Cheers Stéfan

On Tuesday 10 February 2009 11:11:38 Markus Rosenstihl wrote:

i usually do something like this:

a = random.rand(3000) a.resize((1000,3)) vec_norms = sqrt(sum(a**2,axis=1))

If you look at the patch I posted (OK, that was some weeks ago, so I'll attach it again for your convenience), that's (more or less) exactly what I proposed.

Have a nice day, Hans