[Numpy-discussion] ufunc oddities

Charles R Harris charlesr.harris@gmail....
Sun May 25 13:28:06 CDT 2008


On Sun, May 25, 2008 at 11:30 AM, Anne Archibald <peridot.faceted@gmail.com>
wrote:

> 2008/5/25 Charles R Harris <charlesr.harris@gmail.com>:
>
> > So, please tell me how numpy is supposed to work. Write as much as you
> > please. If you are so moved, why not write the tests for all 64 ufuncs
> for
> > all types and combinations and verify that they are all correct as
> specified
> > and raise errors when they should.
>
> This sounds like something that will be easier with nose (if I
> understand nose correctly). Since the ufuncs are so uniform,


The loops are easy enough, the needed information is in the ufuncs:
nin,nout, etc. It is the results that need to be specified because they
aren't, in fact, uniform. And all the promotion rules and such also need to
be specified and checked. So we need to know what is supposed to happen and
pinning that down is the problem.

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://projects.scipy.org/pipermail/numpy-discussion/attachments/20080525/5a5702ee/attachment.html 


More information about the Numpy-discussion mailing list