Dear fenics-team, 

i'm new to fenics. i'm student an working on my
masterthesis, in which i should use fenics. 

>From the page


http://www.dolfin-adjoint.org/en/latest/documentation/optimisation.html


i want to test the example code of the heat equation. There i get
some errors. Please help me to remove them! 

At first in the line (n =
256 for all following lines) 

mesh = Rectangle(-1,-1,1,1,n,n) 

i get
that rectangle is not defined. So i've change it into 

mesh =
RectangleMesh(-1,-1,1,1,n,n) 

and then i get the error message 

File
"/home/chla2091/workspace/test/hitze.py", line 12, in 
 mesh =
RectangleMesh(-1,-1,1,1,n, n)
 File
"/usr/python278/lib/python2.7/site-packages/dolfin/cpp/mesh.py", line
9690, in __init__
 _mesh.RectangleMesh_swiginit(self,
_mesh.new_RectangleMesh(*args))
TypeError: in method
'new_RectangleMesh', argument 1 of type 'MPI_Comm' 

My idea was, only
for testing, to change this line again into 

mesh = UnitSquareMesh(n,
n) 

so it start running with some messages "...this may take some time"
and i think for testing it's ok...but i want also to use the
rectanglemesh in code...it does not work...what i do wrong in this line?


After running there is the next error in line 

m_opt =
minimize(J_Tilde, method = "L-BFGS-B", tol=2e-08, bounds = (0, 1),
options = {"disp": True}) 

No Jacobian form specified for nonlinear
variational problem.
Differentiating residual form F to obtain Jacobian
J = F'.
Solving nonlinear variational problem.
 Newton iteration 0: r
(abs) = 0.000e+00 (tol = 1.000e-10) r (rel) = -nan (tol = 1.000e-09)

Newton solver finished in 0 iterations and 0 linear solver
iterations.
cell.x is deprecated, please use
SpatialCoordinate(domain) instead
RUNNING THE L-BFGS-B CODE

 * *
*

Machine precision = 2.220D-16
 N = 66049 M = 10

At X0 66049
variables are exactly at the bounds
Traceback (most recent call last):

File "/home/chla2091/workspace/test/hitze.py", line 33, in 
 m_opt =
minimize(J_Tilde, method = "L-BFGS-B", tol=2e-08, bounds = (0, 1),
options = {"disp": True})
 File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/optimization/optimization.py",
line 224, in minimize
 opt = algorithm(rf_np, **kwargs)
 File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/optimization/optimization.py",
line 126, in minimize_scipy_generic
 res = scipy_minimize(J, m_global,
method=method, bounds=bounds, **kwargs)
 File
"/usr/python278/lib/python2.7/site-packages/scipy/optimize/_minimize.py",
line 427, in minimize
 callback=callback, **options)
 File
"/usr/python278/lib/python2.7/site-packages/scipy/optimize/lbfgsb.py",
line 314, in _minimize_lbfgsb
 f, g = func_and_grad(x)
 File
"/usr/python278/lib/python2.7/site-packages/scipy/optimize/lbfgsb.py",
line 265, in func_and_grad
 f = fun(x, *args)
 File
"/usr/python278/lib/python2.7/site-packages/scipy/optimize/optimize.py",
line 281, in function_wrapper
 return function(*(wrapper_args + args))

File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/reduced_functional_numpy.py",
line 53, in __call__
 return self.__base_call__(m)
 File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/reduced_functional.py",
line 145, in __call__
 ListControl(self.controls).update(value)
 File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/controls.py",
line 506, in update
 [c.update(v) for c, v in zip(self.controls,
value)]
 File
"/usr/python278/lib/python2.7/site-packages/dolfin_adjoint/controls.py",
line 147, in update
 eqn_nb =
self.var.equation_nb(adjglobals.adjointer)
 File
"/usr/python278/lib/python2.7/site-packages/libadjoint/libadjoint.py",
line 124, in equation_nb

clib.adj_find_variable_equation_nb(adjointer.adjointer, self.var,
equation_nb)
 File
"/usr/python278/lib/python2.7/site-packages/libadjoint/libadjoint.py",
line 18, in handle_error
 raise exception,
errstr
libadjoint.exceptions.LibadjointErrorHashFailed: The queried
variable Control:0:0:Forward is not known to the adjointer. 

So i can't
see plots and solve the problem. Please help me! What is wrong? 

Best
regards and hope to read soon. 
-- 

Christian Landrock 

Student 

 
_______________________________________________
fenics-support mailing list
[email protected]
http://fenicsproject.org/mailman/listinfo/fenics-support

Reply via email to