[Numpy-discussion] 0/0 is 0 or nan?

[Christopher Barker]

Robert Kern wrote:
> However, I believe that divide() on two Python ints goes through a different
> path than for numpy arrays and numpy scalars. Rather, they are recognized as
> objects that implement .__div__(), and that is called instead.

Then wouldn't this be different:

 >>> a = N.array(0, dtype=N.int)
 >>> b = N.array(0, dtype=N.int)
 >>> N.divide(a,b)
0

or to really make sure we're not using Python integers:

 >>> a = N.array(0, dtype=N.int16)
 >>> b = N.array(0, dtype=N.int16)
 >>> N.divide(a,b)
0

for completeness:
 >>> a = N.array(0, dtype=N.float)
 >>> b = N.array(0, dtype=N.float)
 >>> N.divide(a,b)
Warning: invalid value encountered in divide
nan

-Chris



-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker at noaa.gov