Hi, Matthias. We were talking about Steven Watanabe's compile-time linear solver, currently part of Boost.Units. Can you point me in the right direction as to where the implementation is? (My initial guess is <boost/units/detail/linear_algebra.hpp>.) Documentation would be a big plus, but at least I can read source code. Cheers, Cromwell D. Enage
AMDG Cromwell Enage wrote:
We were talking about Steven Watanabe's compile-time linear solver, currently part of Boost.Units. Can you point me in the right direction as to where the implementation is? (My initial guess is <boost/units/detail/linear_algebra.hpp>.)
Yes.
Documentation would be a big plus, but at least I can read source code.
This is not documented since it is implementation details. To solve the matrix equation Ax = b, the code primarily uses Gauss-Jordan elimination to invert A, although there is a special case when A = I. (The assumption is that A is likely the be used several times, so we take advantage of memoization). In Christ, Steven Watanabe
on Thu Nov 13 2008, Steven Watanabe <watanabesj-AT-gmail.com> wrote:
AMDG
Cromwell Enage wrote:
We were talking about Steven Watanabe's compile-time linear solver, currently part of Boost.Units. Can you point me in the right direction as to where the implementation is? (My initial guess is <boost/units/detail/linear_algebra.hpp>.)
Yes.
Documentation would be a big plus, but at least I can read source code.
This is not documented since it is implementation details.
To solve the matrix equation Ax = b, the code primarily uses Gauss-Jordan elimination to invert A, although there is a special case when A = I. (The assumption is that A is likely the be used several times, so we take advantage of memoization).
I just wanna say, that's ridiculous. I mean ridiculously cool, of course. -- Dave Abrahams BoostPro Computing http://www.boostpro.com
participants (3)
-
Cromwell Enage -
David Abrahams -
Steven Watanabe