From: Jeff Garland (jeff_at_[hidden])
Date: 2001-03-09 13:07:20
Take a look at the thread on "N-dimensional generic resizable array" and the
preliminary submission at
http://groups.yahoo.com/group/boost/files/multi_array/. Giovanni Bavestrelli
put this together for a CUJ article which prompted a long discussion of
requirements. It's been pretty quiet for awhile now....
> -----Original Message-----
> From: tk_at_[hidden]
> [mailto:tk_at_[hidden]]On Behalf Of Toon Knapen
> Sent: Friday, March 09, 2001 11:44 AM
> To: boost
> Subject: [boost] matrix library ?
> Would there be any interest in developing a multi-dimensional array
> (vector, matrix, ...) library at Boost ?
> The intent would be to have an easy to use interface and
> high-performance implementation for use in scientific applications or
> whereever multi-dimensional data structure need to be manipulated.
> Following features would at least have to be in there:
> _ templated value_type
> _ Expression Templates for performant and easy interface (e.g. `y = A *
> x + b` without tmp's)
> _ dense and sparse matrices
> _ support for slices / subarrays
> _ common linear algebra operations (addition, multiplication of
> matrices, ....)
> There are already some libraries doing all/some of the above but I'm
> sure that the boost development process and development approach is able
> to produce a next level of matrix libary with respect to quality,
> portability and functionality.
> This way, hopefully some kind of standard can be realised (you hear the
> whole time of new matrix libaries popping up, again and again) such that
> there's totally no excuse anymore (even for scientists) to stay with
> Fortran 77/90.
> List-Unsubscribe: <mailto:boost-unsubscribe_at_[hidden]>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk