Dear colleagues, sorry for hijacking the thread, but could you explain to me, how to compute residue of adjoint functional, if possible, with firedrake-adjoint?

I've seen in the documentation that

J = Functional(inner(u, u)*dx*dt[FINISH_TIME])
for (variable, solution) in compute_adjoint(J):

provides the solution of adjoint problem, but for residual I need not only solution, but also a set of adjoint equations.

Sincerely,
George

On 10.04.2017 17:32, Buesing, Henrik wrote:

Yes! I opened a github issue here. I guess the markdown messes with the python code. Sorry, for that.

Henrik

 

 

--

Dipl.-Math. Henrik Büsing

Institute for Applied Geophysics and Geothermal Energy

E.ON Energy Research Center

RWTH Aachen University

 

Mathieustr.           | Tel +49 (0)241 80 49907

52074 Aachen, Germany | Fax +49 (0)241 80 49889

 

http://www.eonerc.rwth-aachen.de/GGE

hbuesing@eonerc.rwth-aachen.de

 

Von: firedrake-bounces@imperial.ac.uk [mailto:firedrake-bounces@imperial.ac.uk] Im Auftrag von David Ham
Gesendet: Montag, 10. April 2017 11:57
An: firedrake <firedrake@imperial.ac.uk>
Betreff: Re: [firedrake] firedrake_adjoint

 

Hi Henrik,

 

That sounds like a bug. Can you create a minimal failing example and file a github issue please?

 

Regards,

 

David

 

On Mon, 10 Apr 2017 at 10:48 Buesing, Henrik <HBuesing@eonerc.rwth-aachen.de> wrote:

Dear all,

 

If I include

 

from firedrake_adjoint import *

 

in my firedrake program, my forward problem does not converge any more. If I do nothing else than loading firedrake_adjoint, should the forward problem not run as before? Note, that I set all the solver options from the command line and not in the program itself. Thus, in the program I just have

 

solver = NonlinearVariationalSolver(problem,options_prefix="")

Thank you!
Henrik

 

 

--

Dipl.-Math. Henrik Büsing

Institute for Applied Geophysics and Geothermal Energy

E.ON Energy Research Center

RWTH Aachen University

------------------------------------------------------

Mathieustr. 10            |    Tel +49 (0)241 80 49907

52074 Aachen, Germany     |    Fax +49 (0)241 80 49889

------------------------------------------------------

http://www.eonerc.rwth-aachen.de/GGE

hbuesing@eonerc.rwth-aachen.de

------------------------------------------------------

 

--

Dr David Ham

Department of Mathematics

Imperial College London

_______________________________________________
firedrake mailing list
firedrake@imperial.ac.uk
https://mailman.ic.ac.uk/mailman/listinfo/firedrake