On 2016-01-17 21:23, palik imre wrote:
> Hi all,
>
> It seems that the matrix multiplication in ublas ends up with the
> trivial algorithm.  On my machine, even the following function
> outperforms it for square matrices bigger than 173*173 (by a huge
> margin for matrices bigger than 190*190), while not performing
> considerably worse for smaller matrices:
>
> matrix<double>
> matmul_byrow(const matrix<double> &lhs, const matrix<double> &rhs)
> {
>  assert(lhs.size2() == rhs.size1());
>  matrix<double> rv(lhs.size1(), rhs.size2());
>  matrix<double> r = trans(rhs);
>  for (unsigned c = 0; c < rhs.size2(); c++)
>    {
>      matrix_column<matrix<double> > out(rv, c);
>      matrix_row<matrix<double> > in(r, c);
>      out = prod(lhs, in);
>    }
>  return rv;
> }
>
>
> Is there anybody working on improving the matrix multiplication
> performance?
>
> If not, then I can try to find some spare cycles ...
>
> Cheers,
>
> Imre Palik