Can't exponentiate zero in NumPy

Todd Miller jmiller at stsci.edu
Mon Sep 9 20:01:08 CEST 2002


Rick Muller wrote:
> I found out more about this problem, and I'm still hoping that someone 
> can suggest a good fix.
> 
> The problem comes when I take a power of a NumPy array that has very 
> small values in it:
> 
>  >>> from Numeric import *
>  >>> a = zeros(3,Float)
>  >>> a**2
> array([0., 0., 0.])
>  >>> a[1] = 1.e-310
>  >>> a**2
> Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
> OverflowError: math range error
> 
> This crashes on Linux, but not on Mac OS X.
> 
> Is this a bug or a feature?
> 
> If it's a feature (which it might be), can someone offer a one-line 
> workaround? What I would like to do is set everything below a small 
> value (say 1.e-16) to zero. Obviously,
>  >>> for i in range(len(a)): if a[i] < 1.e-16: a[i] = 0
> or something like it should work, but it will also be slow, and I was 
> hoping that one of the Numeric UFuncs might do the same thing more 
> effectively.
> 
> Thanks in advance for any help anyone can offer.
> 
> Rick
> 

I think what you're looking for is "where".  You use it like:

 >>> b = where( abs(a) < 1.e-16, 0, a)

Which means "where the absolute value of a is less than 1.e-16, use 0, 
else use a".  The "<" operator only works like this for Python-2.1 and 
up;  prior to that you have to substitute a "lessthan" function call.


Todd




More information about the Python-list mailing list