[SciPy-Dev] numba simple example

[Dieter Werthmüller]

Just as information, might be useful.

I remember that it was discussed something over a year ago whether or 
not having numba as a dependency. I think it was Ralf who said that it 
would be nice if one could achieve with a simple decorator a speed-up 
with numba - if it is installed, and otherwise just keep the status quo.

The following simple wrap resulted in a speed-up of a factor 2; the 
speed up was pretty constant (in my short tests) for arrays say from a 
couple of ten-thousands to 1e9 complex elements; for smaller array the 
gain was even higher.

@nb.njit(nogil=True, fastmath=True, cache=True)
def l2norm(x):
     """Jitted version of np.linalg.norm(x, ord=None)."""
     return np.linalg.norm(x)

Again, just for the record, for future discussions on the topic.