March 2016 - firedrake - imperial.ac.uk

L-shape mesh from gmsh and boundary conditions
by George Ovchinnikov 31 Mar '16

31 Mar '16

Dear All, I have the following mesh from gmsh (see attached). I can read it just fine: mesh = Mesh("lshape.msh") but when I try to solve equation on this mesh it fails to apply boundary conditions (because boundaries isn't marked in the msh file, I guess): --------------------------------------------------------------------------- LookupError Traceback (most recent call last) <ipython-input-17-ae4ad3326f99> in <module>() 6 fun0 = None 7 cff = 0.002 ----> 8 sol = get_sol(frm, rhs) 9 fun0 = get_functional(sol, f) <ipython-input-15-e7cda2e05bae> in get_sol(frm, rhs) 4 sol = Function(V) 5 #solve(frm, sol, rhs, solver_parameters={'ksp_type': 'cg'}) ----> 6 solve(frm, sol, rhs, solver_parameters={"ksp_type": "preonly", "pc_type": "lu"}) 7 8 return sol /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc in solve(*args, **kwargs) 121 else: 122 # Solve pre-assembled system --> 123 return _la_solve(*args, **kwargs) 124 125 /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc in _la_solve(A, x, b, **kwargs) 209 solver = ls.LinearSolver(A, solver_parameters=solver_parameters, 210 nullspace=nullspace, --> 211 options_prefix=options_prefix) 212 213 solver.solve(x, b) /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/linear_solver.pyc in __init__(self, A, P, solver_parameters, nullspace, options_prefix) 87 # applied 88 # Force evaluation here ---> 89 self.ksp.setOperators(A=self.A.M.handle, P=self.P.M.handle) 90 91 @cached_property /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/matrix.pyc in M(self) 143 just need a handle on the :class:`pyop2.Mat` object it's 144 wrapping, use :attr:`_M` instead.""" --> 145 self.assemble() 146 # User wants to see it, so force the evaluation. 147 self._M._force_evaluation() <decorator-gen-414> in assemble(self) /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/utils.pyc in wrapper(f, *args, **kwargs) 60 opts["type_check"] = safe 61 try: ---> 62 return f(*args, **kwargs) 63 finally: 64 opts["type_check"] = check /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/matrix.pyc in assemble(self) 72 return 73 self._bcs_at_point_of_assembly = copy.copy(self.bcs) ---> 74 self._assembly_callback(self.bcs) 75 self._assembled = True 76 /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/assembly_cache.pyc in inner(bcs) 356 return r 357 --> 358 r = thunk(bcs) 359 if isinstance(r, float): 360 # 0-form case /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/assemble.pyc in thunk(bcs) 328 tensor_arg = mat(lambda s: get_map(s, tsbc, decoration), 329 lambda s: get_map(s, trbc, decoration), --> 330 i, j) 331 elif is_vec: 332 tensor_arg = vec(lambda s: get_map(s), i) /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/assemble.pyc in mat(testmap, trialmap, i, j) 189 def mat(testmap, trialmap, i, j): 190 return tensor[i, j](op2.INC, --> 191 (testmap(test.function_space()[i])[op2.i[0]], 192 trialmap(trial.function_space()[j])[op2.i[1]]), 193 flatten=True) /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/assemble.pyc in <lambda>(s) 326 # Output argument 327 if is_mat: --> 328 tensor_arg = mat(lambda s: get_map(s, tsbc, decoration), 329 lambda s: get_map(s, trbc, decoration), 330 i, j) /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/assemble.pyc in get_map(x, bcs, decoration) 280 281 def get_map(x, bcs=None, decoration=None): --> 282 return x.cell_node_map(bcs) 283 284 elif integral_type in ("exterior_facet", "exterior_facet_vert"): /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/functionspaceimpl.pyc in cell_node_map(self, bcs) 411 "cell_node", 412 self.offset, --> 413 parent) 414 415 def interior_facet_node_map(self, bcs=None): /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/functionspacedata.pyc in get_map(self, V, entity_set, map_arity, bcs, name, offset, parent) 485 # Cache miss. 486 # Any top and bottom bcs (for the extruded case) are handled elsewhere. --> 487 nodes = [bc.nodes for bc in lbcs if bc.sub_domain not in ['top', 'bottom']] 488 decorate = False 489 if nodes: /home/george/polygon/firedrake/local/lib/python2.7/site-packages/pyop2/utils.pyc in __get__(self, obj, cls) 62 if obj is None: 63 return self ---> 64 obj.__dict__[self.__name__] = result = self.fget(obj) 65 return result 66 /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/bcs.pyc in nodes(self) 152 fs.exterior_facet_boundary_node_map( 153 self.method).values_with_halo.take( --> 154 fs._mesh.exterior_facets.subset(self.sub_domain).indices, 155 axis=0)) 156 /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/mesh.pyc in subset(self, markers) 125 raise LookupError( 126 '{0} is not a valid marker'. --> 127 format(marker)) 128 129 # build a list of indices corresponding to the subsets selected by LookupError: 4 is not a valid marker So, I need to somehow mark boundaries, but I cannot find it out, how. I've tried https://openfoamwiki.net/index.php/2D_Mesh_Tutorial_using_GMSH but step 4 does not work for me -- I simply cannot select line with gui, and it is not clear how to select boundaries editing mesh file. Sincerely, George

2 1

Re: [firedrake] Solving adjoint equation
by Andrew McRae 30 Mar '16

30 Mar '16

The matrix appears to be singular, though I don't know why I don't get a zero-pivot error or similar. The code I run is as follow, using 8x8 rather than 64x64 for sanity. --- from firedrake import * size = 8 mesh = UnitSquareMesh(size, size, quadrilateral = True) V = FunctionSpace(mesh, "CG", 1) f = Function(V) f.interpolate(Expression("1")) u_tilde = Function(V) u_tilde.interpolate(Expression("1.0/(1+exp(-10*x[0]))/(1+exp(-10*(1-x[0])))/(1+exp(-10*x[1]))/(1+exp(-10*(1-x[1])))")) w_tilde = TrialFunction(V) v = TestFunction(V) a_tilde = (w_tilde*dot(grad(u_tilde), grad(v)))*dx L = f*v*dx A = assemble(a_tilde) vals = A.M.handle[:, :] import numpy as np print np.linalg.svd(vals)[1] assert False --- This gives me a O(10^-16) singular value, i.e. the matrix is singular. I suspect this is due to the symmetry of u_tilde: grad(u_tilde) is 0 at the middle of the domain, so perhaps that test function is essentially unused. On 30 March 2016 at 11:48, George Ovchinnikov <lives9(a)gmail.com> wrote: > Yes, I think it was connection issue, now I've updated firedrake and > following code > > size = 64 > mesh = UnitSquareMesh(size, size, quadrilateral = True) > V = FunctionSpace(mesh, "CG", 1) > > f = Function(V) > f.interpolate(Expression("1")) > > u_tilde = Function(V) > > u_tilde.interpolate(Expression("1.0/(1+exp(-10*x[0]))/(1+exp(-10*(1-x[0])))/(1+exp(-10*x[1]))/(1+exp(-10*(1-x[1])))" > )) > > w_tilde = TrialFunction(V) > v = TestFunction(V) > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > frm_tilde = assemble( a_tilde ) > > > L = f * v * dx > rhs = assemble(L) > > w_sol = Function(V) > solve(a_tilde == L, w_sol) > > File("w_sol.pvd").write(w_sol) > > runs, but result looks suspicious: the f and u_tilde have symmetries, > while w_sol, does not (look at lower left corner of the image attached). > Also, solve runs only without any parameters, attempt to set any results > in errors. For example > > solve(a_tilde == L, w_sol, solver_parameters={'ksp_type': 'richardson', > 'pc_type': 'jacobi'}) > > results in > > --------------------------------------------------------------------------- > RuntimeError Traceback (most recent call last) > <ipython-input-12-682f5be8ad08> in <module>() > 22 w_sol = Function(V) > 23 solve(a_tilde == L, w_sol, solver_parameters={'ksp_type': > 'richardson', > ---> 24 'pc_type': 'jacobi'}) > 25 > 26 File("w_sol.pvd").write(w_sol) > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in solve(*args, **kwargs) > 118 # Call variational problem solver if we get an equation > 119 if isinstance(args[0], ufl.classes.Equation): > --> 120 _solve_varproblem(*args, **kwargs) > 121 else: > 122 # Solve pre-assembled system > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in _solve_varproblem(*args, **kwargs) > 145 > options_prefix=options_prefix) > 146 with progress(INFO, 'Solving linear variational problem'): > --> 147 solver.solve() > 148 > 149 # Solve nonlinear variational problem > > <decorator-gen-418> in solve(self) > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/pyop2/profiling.pyc > in wrapper(f, *args, **kwargs) > 201 self.stop() > 202 else: > --> 203 return f(*args, **kwargs) > 204 return decorator(wrapper, f) > 205 > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/variational_solver.pyc > in solve(self) > 188 self.snes.solve(None, v) > 189 > --> 190 solving_utils.check_snes_convergence(self.snes) > 191 > 192 > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving_utils.pyc > in check_snes_convergence(snes) > 60 raise RuntimeError("""Nonlinear solve failed to converge > after %d nonlinear iterations. > 61 Reason: > ---> 62 %s""" % (snes.getIterationNumber(), msg)) > 63 > 64 > > RuntimeError: Nonlinear solve failed to converge after 0 nonlinear > iterations. > Reason: > Inner linear solve failed to converge after 3 iterations with reason: > DIVERGED_DTOL > > > > and > solve(a_tilde == L, w_sol, solver_parameters={'ksp_type': 'preonly', > 'pc_type': 'lu'}) > > results in: > > --------------------------------------------------------------------------- > RuntimeError Traceback (most recent call last) > <ipython-input-13-7297871754fa> in <module>() > 22 w_sol = Function(V) > 23 solve(a_tilde == L, w_sol, solver_parameters={'ksp_type': > 'preonly', > ---> 24 'pc_type': 'lu'}) > 25 > 26 File("w_sol.pvd").write(w_sol) > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in solve(*args, **kwargs) > 118 # Call variational problem solver if we get an equation > 119 if isinstance(args[0], ufl.classes.Equation): > --> 120 _solve_varproblem(*args, **kwargs) > 121 else: > 122 # Solve pre-assembled system > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in _solve_varproblem(*args, **kwargs) > 145 > options_prefix=options_prefix) > 146 with progress(INFO, 'Solving linear variational problem'): > --> 147 solver.solve() > 148 > 149 # Solve nonlinear variational problem > > <decorator-gen-418> in solve(self) > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/pyop2/profiling.pyc > in wrapper(f, *args, **kwargs) > 201 self.stop() > 202 else: > --> 203 return f(*args, **kwargs) > 204 return decorator(wrapper, f) > 205 > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/variational_solver.pyc > in solve(self) > 188 self.snes.solve(None, v) > 189 > --> 190 solving_utils.check_snes_convergence(self.snes) > 191 > 192 > > > /home/george/polygon/firedrake/local/lib/python2.7/site-packages/firedrake/solving_utils.pyc > in check_snes_convergence(snes) > 60 raise RuntimeError("""Nonlinear solve failed to converge > after %d nonlinear iterations. > 61 Reason: > ---> 62 %s""" % (snes.getIterationNumber(), msg)) > 63 > 64 > > RuntimeError: Nonlinear solve failed to converge after 0 nonlinear > iterations. > Reason: > Inner linear solve failed to converge after 0 iterations with reason: > unknown reason (petsc4py enum incomplete?) > > > > On 30.03.2016 12:46, Andrew McRae wrote: > > The second and third of those links work for me, so perhaps there's an > > issue at your end. > > > > Anyway, let's bypass this for now by running in "minimal PETSc" mode. > > Can you edit your firedrake-update script (located at, I guess, > > /home/george/polygon/firedrake/bin/firedrake-update) by changing > > args.minimal_petsc from False to True (around line 120-130) and rerun? > > This should install PETSc without most third-party packages, including > > the one you're having trouble with, when you run firedrake-update. > > > > On 29 March 2016 at 20:18, George Ovchinnikov <lives9(a)gmail.com > > <mailto:lives9@gmail.com>> wrote: > > > > I'm trying to update firedrake, but getting > > > > TESTING: configureLibrary from > > > config.packages.netcdf(/tmp/pip-YAhIai-build/config/BuildSystem/config/package.py:626) > > > > > =============================================================================== > > Trying to download > > ftp://ftp.unidata.ucar.edu/pub/netcdf/netcdf-4.3.2.tar.gz for NETCDF > > > > > =============================================================================== > > > > > > > =============================================================================== > > Trying to download > > > http://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz > > for NETCDF > > > > > =============================================================================== > > > > > > > =============================================================================== > > Trying to download > > ftp://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz > for > > NETCDF > > > > > =============================================================================== > > > > > > > ******************************************************************************* > > UNABLE to CONFIGURE with GIVEN OPTIONS (see > > configure.log for details): > > > > > ------------------------------------------------------------------------------- > > Unable to download NETCDF > > Unable to download package NETCDF from: > > ftp://ftp.unidata.ucar.edu/pub/netcdf/netcdf-4.3.2.tar.gz > > * If URL specified manually - perhaps there is a typo? > > * If your network is disconnected - please reconnect and rerun > > ./configure > > * Or perhaps you have a firewall blocking the download > > * Alternatively, you can download the above URL manually, to > > /yourselectedlocation/netcdf-4.3.2.tar.gz > > and use the configure option: > > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > > file could not be opened successfully > > Downloaded package NETCDF from: > > > http://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz is > > not a tarball. > > [or installed python cannot process compressed files] > > * If you are behind a firewall - please fix your proxy and rerun > > ./configure > > For example at LANL you may need to set the environmental > variable > > http_proxy (or HTTP_PROXY?) to http://proxyout.lanl.gov > > * Alternatively, you can download the above URL manually, to > > /yourselectedlocation/netcdf-4.3.2.tar.gz > > and use the configure option: > > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > > Unable to download package NETCDF from: > > ftp://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz > > * If URL specified manually - perhaps there is a typo? > > * If your network is disconnected - please reconnect and rerun > > ./configure > > * Or perhaps you have a firewall blocking the download > > * Alternatively, you can download the above URL manually, to > > /yourselectedlocation/netcdf-4.3.2.tar.gz > > and use the configure option: > > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > > > > > ******************************************************************************* > > > > Traceback (most recent call last): > > File "<string>", line 1, in <module> > > File "/tmp/pip-YAhIai-build/setup.py", line 302, in <module> > > **metadata) > > File "/usr/lib/python2.7/distutils/core.py", line 151, in setup > > dist.run_commands() > > File "/usr/lib/python2.7/distutils/dist.py", line 953, in > > run_commands > > self.run_command(cmd) > > File "/usr/lib/python2.7/distutils/dist.py", line 972, in > > run_command > > cmd_obj.run() > > File "/tmp/pip-YAhIai-build/setup.py", line 218, in run > > config(prefix, self.dry_run) > > File "/tmp/pip-YAhIai-build/setup.py", line 148, in config > > if status != 0: raise RuntimeError(status) > > RuntimeError: 256 > > > > ---------------------------------------- > > Command "/home/george/polygon/firedrake/bin/python -u -c "import > > setuptools, > > > tokenize;__file__='/tmp/pip-YAhIai-build/setup.py';exec(compile(getattr(tokenize, > > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, > > 'exec'))" install --record /tmp/pip-OhlEEH-record/install-record.txt > > --single-version-externally-managed --compile --install-headers > > /home/george/polygon/firedrake/include/site/python2.7/petsc" failed > with > > error code 1 in /tmp/pip-YAhIai-build/ > > Traceback (most recent call last): > > File "firedrake-install", line 836, in <module> > > install("petsc/") > > File "firedrake-install", line 497, in install > > run_pip_install(["--ignore-installed", package]) > > File "firedrake-install", line 360, in run_pip_install > > check_call(pipinstall + pipargs) > > File "firedrake-install", line 211, in check_call > > subprocess.check_call(arguments, env=env) > > File "/usr/lib/python2.7/subprocess.py", line 540, in check_call > > raise CalledProcessError(retcode, cmd) > > subprocess.CalledProcessError: Command > > '['/home/george/polygon/firedrake/bin/pip', 'install', '--no-deps', > > '--ignore-installed', 'petsc/']' returned non-zero exit status 1 > > > > > > On all of the abovementioned urls I'm getting 403 error when trying > to > > access manually. > > > > Sincerely, > > George > > > > > > On 29.03.2016 18:30, Andrew McRae wrote: > > > I also note that your line numbers don't match up with the current > > > codebase; e.g. > > > > https://github.com/firedrakeproject/firedrake/blob/master/firedrake/linear_… > > > > > > On 29 March 2016 at 16:13, George Ovchinnikov <lives9(a)gmail.com > <mailto:lives9@gmail.com> > > > <mailto:lives9@gmail.com <mailto:lives9@gmail.com>>> wrote: > > > > > > Here it is > > > > > > > > > --------------------------------------------------------------------------- > > > KeyError Traceback (most > recent > > > call last) > > > <ipython-input-8-752634385ffb> in <module>() > > > 22 > > > 23 w_sol = Function(V) > > > ---> 24 solve(frm_tilde, rhs, w_sol, > > solver_parameters={"ksp_type": > > > "preonly", "pc_type": "lu"}) > > > 25 File("w_sol.pvd") << w_sol > > > > > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > > > in solve(*args, **kwargs) > > > 121 else: > > > 122 # Solve pre-assembled system > > > --> 123 return _la_solve(*args, **kwargs) > > > 124 > > > 125 > > > > > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > > > in _la_solve(A, x, b, **kwargs) > > > 209 > > options_prefix=options_prefix) > > > 210 > > > --> 211 solver.solve(x, b) > > > 212 > > > 213 > > > > > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/linear_solver.pyc > > > in solve(self, x, b) > > > 140 r = self.ksp.getConvergedReason() > > > 141 if r < 0: > > > --> 142 raise RuntimeError("LinearSolver failed to > > converge > > > after %d iterations with reason: %s", > > self.ksp.getIterationNumber(), > > > solving_utils.KSPReasons[r]) > > > > > > KeyError: -11 > > > > > > > > > > > > On 29.03.2016 18:10, Andrew McRae wrote: > > > > What output do you get? > > > > > > > > On 29 March 2016 at 16:09, George Ovchinnikov > > <lives9(a)gmail.com <mailto:lives9@gmail.com> <mailto:lives9@gmail.com > > <mailto:lives9@gmail.com>> > > > > <mailto:lives9@gmail.com <mailto:lives9@gmail.com> > > <mailto:lives9@gmail.com <mailto:lives9@gmail.com>>>> wrote: > > > > > > > > Dear All, > > > > > > > > I have trilinear form a(w,u,v) and usually solve > > > > a(w,u,v) = l(v), for u with fixed w. > > > > > > > > Here I have: > > > > > > > > a = w dot(grad(u), grad(v)) * dx, > > > > and > > > > l(v) = f * v * dx. > > > > > > > > Now, given u_tilde I want to solve for w, i.e. find > > > > w_tilde such, that > > > > a(w_tilde, u_tilde, v) = l(v), for any v in suitable > > subspace. > > > > > > > > I'm trying the following: > > > > > > > > mesh = UnitSquareMesh(size, size) > > > > V = FunctionSpace(mesh, "CG", 1) > > > > > > > > f = Function(V) > > > > f.interpolate(Expression("1")) > > > > > > > > u_tilde = Function(V) > > > > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > > > > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * > > > 1/(1+exp(-1*(1-x[1])))" )) > > > > > > > > w_tilde = TrialFunction(V) > > > > v = TestFunction(V) > > > > > > > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > > > > frm_tilde = assemble( a_tilde ) > > > > > > > > L = f * v * dx > > > > rhs = assemble(L) > > > > > > > > w_sol = Function(V) > > > > solve(frm_tilde, rhs, w_sol, > solver_parameters={"ksp_type": > > > "preonly", > > > > "pc_type": "lu"}) > > > > File("w_sol.pvd") << w_sol > > > > > > > > > > > > It fails to converge: > > > > 140 r = self.ksp.getConvergedReason() > > > > 141 if r < 0: > > > > --> 142 raise RuntimeError("LinearSolver > > failed to > > > converge > > > > after %d iterations with reason: %s", > > > self.ksp.getIterationNumber(), > > > > solving_utils.KSPReasons[r]) > > > > > > > > KeyError: -11 > > > > > > > > I'm using direct method, not the iterative one, so what > is > > > happening > > > > here? > > > > > > > > _______________________________________________ > > > > firedrake mailing list > > > > firedrake(a)imperial.ac.uk > > <mailto:firedrake@imperial.ac.uk> <mailto:firedrake@imperial.ac.uk > > <mailto:firedrake@imperial.ac.uk>> > > > <mailto:firedrake@imperial.ac.uk > > <mailto:firedrake@imperial.ac.uk> <mailto:firedrake@imperial.ac.uk > > <mailto:firedrake@imperial.ac.uk>>> > > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > > > > > > > > > > > _______________________________________________ > > > > firedrake mailing list > > > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > <mailto:firedrake@imperial.ac.uk <mailto:firedrake@imperial.ac.uk>> > > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > > > > _______________________________________________ > > > firedrake mailing list > > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > <mailto:firedrake@imperial.ac.uk <mailto:firedrake@imperial.ac.uk>> > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > > > > > > _______________________________________________ > > > firedrake mailing list > > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > >

2 2

Re: [firedrake] Solving adjoint equation
by Andrew McRae 30 Mar '16

30 Mar '16

The second and third of those links work for me, so perhaps there's an issue at your end. Anyway, let's bypass this for now by running in "minimal PETSc" mode. Can you edit your firedrake-update script (located at, I guess, /home/george/polygon/firedrake/bin/firedrake-update) by changing args.minimal_petsc from False to True (around line 120-130) and rerun? This should install PETSc without most third-party packages, including the one you're having trouble with, when you run firedrake-update. On 29 March 2016 at 20:18, George Ovchinnikov <lives9(a)gmail.com> wrote: > I'm trying to update firedrake, but getting > > TESTING: configureLibrary from > > config.packages.netcdf(/tmp/pip-YAhIai-build/config/BuildSystem/config/package.py:626) > > > =============================================================================== > Trying to download > ftp://ftp.unidata.ucar.edu/pub/netcdf/netcdf-4.3.2.tar.gz for NETCDF > > > =============================================================================== > > > > =============================================================================== > Trying to download > http://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz > for NETCDF > > > =============================================================================== > > > > =============================================================================== > Trying to download > ftp://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz for > NETCDF > > > =============================================================================== > > > > ******************************************************************************* > UNABLE to CONFIGURE with GIVEN OPTIONS (see > configure.log for details): > > > ------------------------------------------------------------------------------- > Unable to download NETCDF > Unable to download package NETCDF from: > ftp://ftp.unidata.ucar.edu/pub/netcdf/netcdf-4.3.2.tar.gz > * If URL specified manually - perhaps there is a typo? > * If your network is disconnected - please reconnect and rerun > ./configure > * Or perhaps you have a firewall blocking the download > * Alternatively, you can download the above URL manually, to > /yourselectedlocation/netcdf-4.3.2.tar.gz > and use the configure option: > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > file could not be opened successfully > Downloaded package NETCDF from: > http://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz is > not a tarball. > [or installed python cannot process compressed files] > * If you are behind a firewall - please fix your proxy and rerun > ./configure > For example at LANL you may need to set the environmental variable > http_proxy (or HTTP_PROXY?) to http://proxyout.lanl.gov > * Alternatively, you can download the above URL manually, to > /yourselectedlocation/netcdf-4.3.2.tar.gz > and use the configure option: > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > Unable to download package NETCDF from: > ftp://ftp.mcs.anl.gov/pub/petsc/externalpackages/netcdf-4.3.2.tar.gz > * If URL specified manually - perhaps there is a typo? > * If your network is disconnected - please reconnect and rerun > ./configure > * Or perhaps you have a firewall blocking the download > * Alternatively, you can download the above URL manually, to > /yourselectedlocation/netcdf-4.3.2.tar.gz > and use the configure option: > --download-netcdf=/yourselectedlocation/netcdf-4.3.2.tar.gz > > > ******************************************************************************* > > Traceback (most recent call last): > File "<string>", line 1, in <module> > File "/tmp/pip-YAhIai-build/setup.py", line 302, in <module> > **metadata) > File "/usr/lib/python2.7/distutils/core.py", line 151, in setup > dist.run_commands() > File "/usr/lib/python2.7/distutils/dist.py", line 953, in > run_commands > self.run_command(cmd) > File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command > cmd_obj.run() > File "/tmp/pip-YAhIai-build/setup.py", line 218, in run > config(prefix, self.dry_run) > File "/tmp/pip-YAhIai-build/setup.py", line 148, in config > if status != 0: raise RuntimeError(status) > RuntimeError: 256 > > ---------------------------------------- > Command "/home/george/polygon/firedrake/bin/python -u -c "import > setuptools, > > tokenize;__file__='/tmp/pip-YAhIai-build/setup.py';exec(compile(getattr(tokenize, > 'open', open)(__file__).read().replace('\r\n', '\n'), __file__, > 'exec'))" install --record /tmp/pip-OhlEEH-record/install-record.txt > --single-version-externally-managed --compile --install-headers > /home/george/polygon/firedrake/include/site/python2.7/petsc" failed with > error code 1 in /tmp/pip-YAhIai-build/ > Traceback (most recent call last): > File "firedrake-install", line 836, in <module> > install("petsc/") > File "firedrake-install", line 497, in install > run_pip_install(["--ignore-installed", package]) > File "firedrake-install", line 360, in run_pip_install > check_call(pipinstall + pipargs) > File "firedrake-install", line 211, in check_call > subprocess.check_call(arguments, env=env) > File "/usr/lib/python2.7/subprocess.py", line 540, in check_call > raise CalledProcessError(retcode, cmd) > subprocess.CalledProcessError: Command > '['/home/george/polygon/firedrake/bin/pip', 'install', '--no-deps', > '--ignore-installed', 'petsc/']' returned non-zero exit status 1 > > > On all of the abovementioned urls I'm getting 403 error when trying to > access manually. > > Sincerely, > George > > > On 29.03.2016 18:30, Andrew McRae wrote: > > I also note that your line numbers don't match up with the current > > codebase; e.g. > > > https://github.com/firedrakeproject/firedrake/blob/master/firedrake/linear_… > > > > On 29 March 2016 at 16:13, George Ovchinnikov <lives9(a)gmail.com > > <mailto:lives9@gmail.com>> wrote: > > > > Here it is > > > > > --------------------------------------------------------------------------- > > KeyError Traceback (most recent > > call last) > > <ipython-input-8-752634385ffb> in <module>() > > 22 > > 23 w_sol = Function(V) > > ---> 24 solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > > "preonly", "pc_type": "lu"}) > > 25 File("w_sol.pvd") << w_sol > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > > in solve(*args, **kwargs) > > 121 else: > > 122 # Solve pre-assembled system > > --> 123 return _la_solve(*args, **kwargs) > > 124 > > 125 > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > > in _la_solve(A, x, b, **kwargs) > > 209 options_prefix=options_prefix) > > 210 > > --> 211 solver.solve(x, b) > > 212 > > 213 > > > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/linear_solver.pyc > > in solve(self, x, b) > > 140 r = self.ksp.getConvergedReason() > > 141 if r < 0: > > --> 142 raise RuntimeError("LinearSolver failed to > converge > > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > > solving_utils.KSPReasons[r]) > > > > KeyError: -11 > > > > > > > > On 29.03.2016 18:10, Andrew McRae wrote: > > > What output do you get? > > > > > > On 29 March 2016 at 16:09, George Ovchinnikov <lives9(a)gmail.com > <mailto:lives9@gmail.com> > > > <mailto:lives9@gmail.com <mailto:lives9@gmail.com>>> wrote: > > > > > > Dear All, > > > > > > I have trilinear form a(w,u,v) and usually solve > > > a(w,u,v) = l(v), for u with fixed w. > > > > > > Here I have: > > > > > > a = w dot(grad(u), grad(v)) * dx, > > > and > > > l(v) = f * v * dx. > > > > > > Now, given u_tilde I want to solve for w, i.e. find > > > w_tilde such, that > > > a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. > > > > > > I'm trying the following: > > > > > > mesh = UnitSquareMesh(size, size) > > > V = FunctionSpace(mesh, "CG", 1) > > > > > > f = Function(V) > > > f.interpolate(Expression("1")) > > > > > > u_tilde = Function(V) > > > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > > > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * > > 1/(1+exp(-1*(1-x[1])))" )) > > > > > > w_tilde = TrialFunction(V) > > > v = TestFunction(V) > > > > > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > > > frm_tilde = assemble( a_tilde ) > > > > > > L = f * v * dx > > > rhs = assemble(L) > > > > > > w_sol = Function(V) > > > solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > > "preonly", > > > "pc_type": "lu"}) > > > File("w_sol.pvd") << w_sol > > > > > > > > > It fails to converge: > > > 140 r = self.ksp.getConvergedReason() > > > 141 if r < 0: > > > --> 142 raise RuntimeError("LinearSolver failed to > > converge > > > after %d iterations with reason: %s", > > self.ksp.getIterationNumber(), > > > solving_utils.KSPReasons[r]) > > > > > > KeyError: -11 > > > > > > I'm using direct method, not the iterative one, so what is > > happening > > > here? > > > > > > _______________________________________________ > > > firedrake mailing list > > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > <mailto:firedrake@imperial.ac.uk <mailto:firedrake@imperial.ac.uk>> > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > > > > > > _______________________________________________ > > > firedrake mailing list > > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

2 1

confirm
by G. D. McBain 29 Mar '16

29 Mar '16

confirm 3659d40f27a64e5837a16cdd48c8f9d71f7b4df1

1 0

Re: [firedrake] Solving adjoint equation
by Andrew McRae 29 Mar '16

29 Mar '16

I also note that your line numbers don't match up with the current codebase; e.g. https://github.com/firedrakeproject/firedrake/blob/master/firedrake/linear_… On 29 March 2016 at 16:13, George Ovchinnikov <lives9(a)gmail.com> wrote: > Here it is > > --------------------------------------------------------------------------- > KeyError Traceback (most recent call last) > <ipython-input-8-752634385ffb> in <module>() > 22 > 23 w_sol = Function(V) > ---> 24 solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > "preonly", "pc_type": "lu"}) > 25 File("w_sol.pvd") << w_sol > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in solve(*args, **kwargs) > 121 else: > 122 # Solve pre-assembled system > --> 123 return _la_solve(*args, **kwargs) > 124 > 125 > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in _la_solve(A, x, b, **kwargs) > 209 options_prefix=options_prefix) > 210 > --> 211 solver.solve(x, b) > 212 > 213 > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/linear_solver.pyc > in solve(self, x, b) > 140 r = self.ksp.getConvergedReason() > 141 if r < 0: > --> 142 raise RuntimeError("LinearSolver failed to converge > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > solving_utils.KSPReasons[r]) > > KeyError: -11 > > > > On 29.03.2016 18:10, Andrew McRae wrote: > > What output do you get? > > > > On 29 March 2016 at 16:09, George Ovchinnikov <lives9(a)gmail.com > > <mailto:lives9@gmail.com>> wrote: > > > > Dear All, > > > > I have trilinear form a(w,u,v) and usually solve > > a(w,u,v) = l(v), for u with fixed w. > > > > Here I have: > > > > a = w dot(grad(u), grad(v)) * dx, > > and > > l(v) = f * v * dx. > > > > Now, given u_tilde I want to solve for w, i.e. find > > w_tilde such, that > > a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. > > > > I'm trying the following: > > > > mesh = UnitSquareMesh(size, size) > > V = FunctionSpace(mesh, "CG", 1) > > > > f = Function(V) > > f.interpolate(Expression("1")) > > > > u_tilde = Function(V) > > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * 1/(1+exp(-1*(1-x[1])))" > )) > > > > w_tilde = TrialFunction(V) > > v = TestFunction(V) > > > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > > frm_tilde = assemble( a_tilde ) > > > > L = f * v * dx > > rhs = assemble(L) > > > > w_sol = Function(V) > > solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > "preonly", > > "pc_type": "lu"}) > > File("w_sol.pvd") << w_sol > > > > > > It fails to converge: > > 140 r = self.ksp.getConvergedReason() > > 141 if r < 0: > > --> 142 raise RuntimeError("LinearSolver failed to > converge > > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > > solving_utils.KSPReasons[r]) > > > > KeyError: -11 > > > > I'm using direct method, not the iterative one, so what is happening > > here? > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

2 1

Solving adjoint equation
by George Ovchinnikov 29 Mar '16

29 Mar '16

Dear All, I have trilinear form a(w,u,v) and usually solve a(w,u,v) = l(v), for u with fixed w. Here I have: a = w dot(grad(u), grad(v)) * dx, and l(v) = f * v * dx. Now, given u_tilde I want to solve for w, i.e. find w_tilde such, that a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. I'm trying the following: mesh = UnitSquareMesh(size, size) V = FunctionSpace(mesh, "CG", 1) f = Function(V) f.interpolate(Expression("1")) u_tilde = Function(V) u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * 1/(1+exp(-1*(1-x[1])))" )) w_tilde = TrialFunction(V) v = TestFunction(V) a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx frm_tilde = assemble( a_tilde ) L = f * v * dx rhs = assemble(L) w_sol = Function(V) solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": "preonly", "pc_type": "lu"}) File("w_sol.pvd") << w_sol It fails to converge: 140 r = self.ksp.getConvergedReason() 141 if r < 0: --> 142 raise RuntimeError("LinearSolver failed to converge after %d iterations with reason: %s", self.ksp.getIterationNumber(), solving_utils.KSPReasons[r]) KeyError: -11 I'm using direct method, not the iterative one, so what is happening here?

2 2

Re: [firedrake] Solving adjoint equation
by Andrew McRae 29 Mar '16

29 Mar '16

I see; the syntax should be solve(frm_tilde, w_sol, rhs, solver_parameters=...) i.e. "solve(A, x, b)" rather than "solve(A, b, x)". On 29 March 2016 at 16:09, George Ovchinnikov <lives9(a)gmail.com> wrote: > Dear All, > > I have trilinear form a(w,u,v) and usually solve > a(w,u,v) = l(v), for u with fixed w. > > Here I have: > > a = w dot(grad(u), grad(v)) * dx, > and > l(v) = f * v * dx. > > Now, given u_tilde I want to solve for w, i.e. find > w_tilde such, that > a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. > > I'm trying the following: > > mesh = UnitSquareMesh(size, size) > V = FunctionSpace(mesh, "CG", 1) > > f = Function(V) > f.interpolate(Expression("1")) > > u_tilde = Function(V) > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * 1/(1+exp(-1*(1-x[1])))" )) > > w_tilde = TrialFunction(V) > v = TestFunction(V) > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > frm_tilde = assemble( a_tilde ) > > L = f * v * dx > rhs = assemble(L) > > w_sol = Function(V) > solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": "preonly", > "pc_type": "lu"}) > File("w_sol.pvd") << w_sol > > > It fails to converge: > 140 r = self.ksp.getConvergedReason() > 141 if r < 0: > --> 142 raise RuntimeError("LinearSolver failed to converge > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > solving_utils.KSPReasons[r]) > > KeyError: -11 > > I'm using direct method, not the iterative one, so what is happening here? > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

1 0

Re: [firedrake] Solving adjoint equation
by Andrew McRae 29 Mar '16

29 Mar '16

Do you not get any PETSc output before that? The KeyError is weird; according to http://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/KSP/KSPConverge…, -11 is KSP_DIVERGED_PCSETUP_FAILED. I don't use the preassembly syntax normally; does it work if you remove the lines assembling frm_tilde and rhs, and simply call solve(a_tilde == L, w_sol, solver_parameters={...}) ? On 29 March 2016 at 16:13, George Ovchinnikov <lives9(a)gmail.com> wrote: > Here it is > > --------------------------------------------------------------------------- > KeyError Traceback (most recent call last) > <ipython-input-8-752634385ffb> in <module>() > 22 > 23 w_sol = Function(V) > ---> 24 solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > "preonly", "pc_type": "lu"}) > 25 File("w_sol.pvd") << w_sol > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in solve(*args, **kwargs) > 121 else: > 122 # Solve pre-assembled system > --> 123 return _la_solve(*args, **kwargs) > 124 > 125 > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/solving.pyc > in _la_solve(A, x, b, **kwargs) > 209 options_prefix=options_prefix) > 210 > --> 211 solver.solve(x, b) > 212 > 213 > > > /home/george/polygon/firedrake_new/firedrake/local/lib/python2.7/site-packages/firedrake/linear_solver.pyc > in solve(self, x, b) > 140 r = self.ksp.getConvergedReason() > 141 if r < 0: > --> 142 raise RuntimeError("LinearSolver failed to converge > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > solving_utils.KSPReasons[r]) > > KeyError: -11 > > > > On 29.03.2016 18:10, Andrew McRae wrote: > > What output do you get? > > > > On 29 March 2016 at 16:09, George Ovchinnikov <lives9(a)gmail.com > > <mailto:lives9@gmail.com>> wrote: > > > > Dear All, > > > > I have trilinear form a(w,u,v) and usually solve > > a(w,u,v) = l(v), for u with fixed w. > > > > Here I have: > > > > a = w dot(grad(u), grad(v)) * dx, > > and > > l(v) = f * v * dx. > > > > Now, given u_tilde I want to solve for w, i.e. find > > w_tilde such, that > > a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. > > > > I'm trying the following: > > > > mesh = UnitSquareMesh(size, size) > > V = FunctionSpace(mesh, "CG", 1) > > > > f = Function(V) > > f.interpolate(Expression("1")) > > > > u_tilde = Function(V) > > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * 1/(1+exp(-1*(1-x[1])))" > )) > > > > w_tilde = TrialFunction(V) > > v = TestFunction(V) > > > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > > frm_tilde = assemble( a_tilde ) > > > > L = f * v * dx > > rhs = assemble(L) > > > > w_sol = Function(V) > > solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": > "preonly", > > "pc_type": "lu"}) > > File("w_sol.pvd") << w_sol > > > > > > It fails to converge: > > 140 r = self.ksp.getConvergedReason() > > 141 if r < 0: > > --> 142 raise RuntimeError("LinearSolver failed to > converge > > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > > solving_utils.KSPReasons[r]) > > > > KeyError: -11 > > > > I'm using direct method, not the iterative one, so what is happening > > here? > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk <mailto:firedrake@imperial.ac.uk> > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > > > > > > _______________________________________________ > > firedrake mailing list > > firedrake(a)imperial.ac.uk > > https://mailman.ic.ac.uk/mailman/listinfo/firedrake > > > > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

1 0

Re: [firedrake] Solving adjoint equation
by Andrew McRae 29 Mar '16

29 Mar '16

What output do you get? On 29 March 2016 at 16:09, George Ovchinnikov <lives9(a)gmail.com> wrote: > Dear All, > > I have trilinear form a(w,u,v) and usually solve > a(w,u,v) = l(v), for u with fixed w. > > Here I have: > > a = w dot(grad(u), grad(v)) * dx, > and > l(v) = f * v * dx. > > Now, given u_tilde I want to solve for w, i.e. find > w_tilde such, that > a(w_tilde, u_tilde, v) = l(v), for any v in suitable subspace. > > I'm trying the following: > > mesh = UnitSquareMesh(size, size) > V = FunctionSpace(mesh, "CG", 1) > > f = Function(V) > f.interpolate(Expression("1")) > > u_tilde = Function(V) > u_tilde.interpolate(Expression("1/(1+exp(-1*x[0])) * > 1/(1+exp(-1*(1-x[0]))) *1/(1+exp(-1*x[1])) * 1/(1+exp(-1*(1-x[1])))" )) > > w_tilde = TrialFunction(V) > v = TestFunction(V) > > a_tilde = (w_tilde * dot(grad(u_tilde), grad(v))) * dx > frm_tilde = assemble( a_tilde ) > > L = f * v * dx > rhs = assemble(L) > > w_sol = Function(V) > solve(frm_tilde, rhs, w_sol, solver_parameters={"ksp_type": "preonly", > "pc_type": "lu"}) > File("w_sol.pvd") << w_sol > > > It fails to converge: > 140 r = self.ksp.getConvergedReason() > 141 if r < 0: > --> 142 raise RuntimeError("LinearSolver failed to converge > after %d iterations with reason: %s", self.ksp.getIterationNumber(), > solving_utils.KSPReasons[r]) > > KeyError: -11 > > I'm using direct method, not the iterative one, so what is happening here? > > _______________________________________________ > firedrake mailing list > firedrake(a)imperial.ac.uk > https://mailman.ic.ac.uk/mailman/listinfo/firedrake >

2 1

Projecting function to finer mesh
by George Ovchinnikov 29 Mar '16

29 Mar '16

Dear All, I have a function w: mesh = UnitSquareMesh(size, size, quadrilateral = True) V0 = FunctionSpace(mesh, "DG", 0) w = Function(V0) ... w was calculated during my program run. Now, I want to project it to finer mesh, something like: mesh2 = UnitSquareMesh(2*size, 2*size, quadrilateral = True) V2_0 = FunctionSpace(mesh, "DG", 0) w2 = Function(V2_0) project(w, w2) but I'm getting errors about mesh being quadrilateral or something like RuntimeError: Can't project between mismatching meshes. What is the correct way to project function from one mesh to another? Sincerely, George

3 3