[PYTHON MATRIX-SIG] Incompatible change to dot (any objections?)

[Jim Hugunin]

Konrad Hinsen writes:
> I still think that it would be better to give Numeric.dot() the
> functionality of Numeric.matrixmultiply() and remove the latter (or
> keep it for compatibility, but not document it). There is no need for
> two functions, and many people will probably use dot() although what
> they need is matrixmultiply().

It's amazing to me that this has been an issue since the very beginning of 
the matrix-sig.  The problem seems to be that I have much more training in 
computer languages/systems than in linear algebra which gives me a slightly 
different (warped?) perspective on things.

Anyway, I'm finally willing to go along with Konrad on this one.  I intend 
to change dot in the next release to have the same functionality as 
matrixmultiply currently does.  This will give those linear algebra folks 
out there a nice easy way to multiply two matrices together "correctly".  I 
will also expose the lower level function "innerproduct" that currently 
behaves almost exactly like dot (no transposes involved).  I'd also like to 
get rid of the function "matrixmultiply", though I'd be willing to keep it 
around if people have concerns about backwards compatibility for their 
code.

-Jim




_______________
MATRIX-SIG  - SIG on Matrix Math for Python

send messages to: matrix-sig@python.org
administrivia to: matrix-sig-request@python.org
_______________