[Numpy-discussion] extension to xrange for numpy

Gary Ruben gruben at bigpond.net.au
Sun Apr 2 16:06:10 EDT 2006


Doh!
It's OK Arnd; I've recently seen you (or someone else withe the same 
name) acknowledged in a PhD I've been reading so I suspect you're a nice 
guy :-)

And, thanks Alan.
I knew about mgrid but not ogrid.
One small way in which that example might be better than using ogrid is 
that you could avoid creating the index arrays and lazily generate the 
indices. However, ogrid is better than mgrid in this respect.

thanks,
Gary

Alan G Isaac wrote:
> On Sun, 02 Apr 2006, Gary Ruben apparently wrote: 
>> I'd find some sort of numpy iterator equivalents of these more 
>> useful. This would allow list comprehensions like 
>> [f(x,y,z) for (x,y,z) in ndrange(10,10,10)] 
> 
> How is this better than using ogrid? E.g.,
> 
>>>> x=N.ogrid[:3,:2]
>>>> N.power(*x)
> array([[1, 0],
>        [1, 1],
>        [1, 2]])
> 
> Thanks,
> Alan




More information about the NumPy-Discussion mailing list