[CppAD] sparse jacobians

[bradbell at seanet.com]

> Hey,
>
> I am cleaning up some of my code and was looking into replacing my
> hardcoded jacobians with AD ones (makes everybody's life much easier).
> I was curious if it's possible to build sparse jacobians.
>
> Best,
> Vlad
>
> _______________________________________________
> CppAD mailing list
> CppAD at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/cppad
>

There is support for sparse Jacobians in CppAD. For a quick and easy
introduction see
   http://www.coin-or.org/CppAD/Doc/sparse_jacobian.xml
This uses the sparsity to speed up the calculation of the Jacobian, but it
returns a full matrix.

It has been suggested that for very large problems, a spasre
representation of the Jacobian is also necessary. Would it help your
application if sparse_jacobian was extended to allow for returning sparse
representations of the Jacobian ?

In general, if sparse_jacobian is not doing what you need, you can look at
the source code
     https://projects.coin-or.org/CppAD/browser/trunk/cppad/local/sparse_jacobian.hpp
and perhaps make your own version that does what you need.