Boost logo

Boost :

Subject: Re: [boost] different matrix library?
From: joel (joel.falcou_at_[hidden])
Date: 2009-08-15 02:30:11


Edward Grace wrote:

> The more I think about it the less inclined I am towards 'tensors are
> the one true way'. Since we can ultimately represent everything
> (conceptually) as matrices it may (perversely) make some sense to
> treat tensors as a special type of matrix, embuing them with certain
> properties.
This is right. I think we can have a core representation of "N-dimension
body of numerical data" that get used as the base for all matrix/tensor
class and expression and not the other way around.
> Well, semantics I know but they are all tensors. A
>
> scalar -> tensor of order 0,
> vector -> a tensor of order 1,
> matrix -> a tensor of order 2. ;-)
This is just giving rules to transform one body into the other.
And just because I like repeating myself, vector is IMHO redundant.

-- 
___________________________________________
Joel Falcou - Assistant Professor
PARALL Team - LRI - Universite Paris Sud XI
Tel : (+33)1 69 15 66 35

Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk