[SciPy-dev] Hessenberg form

Fernando Perez Fernando.Perez at colorado.edu
Fri Oct 1 13:05:44 CDT 2004


Nils Wagner schrieb:
> On Fri, 1 Oct 2004 09:20:14 -0400 (Eastern Daylight Time)
>   Alan G Isaac <aisaac at american.edu> wrote:
> 
>>On Fri, 01 Oct 2004, Nils Wagner apparently wrote:
>>
>>>I'd like to have a function called hess
>>>in linalg which reduces a general matrix A to upper 
>>>Hessenberg form
>>>before the next release.
>>>Any comment ?
>>
>>The suggested name is perhaps unfortunate.  I would 
>>expect
>>hess to take a function and point as arguments and to 
>>return
>>a matrix of numerical partial derivatives (i.e., the
>>Hessian).
>>
>>Cheers,
>>Alan Isaac
>>
> 
> BTW, Matlab uses the same naming convention.
> 
> http://www.mathworks.com/access/helpdesk/help/techdoc/ref/hess.html

Which doesn't make it right.

I'd actually much rather have hessian and hessenberg be the actual names, to 
avoid this kind of potential confusion.  Long names may be a bit annoying, but 
in my book clarity tends to win.  Individual users can always do 
'hess=hessenberg' inside their codes if so desired, away from public shame :)

Best,

f




More information about the Scipy-dev mailing list