Skip to content
Snippets Groups Projects

Repository graph

You can move around the graph by using the arrow keys.
Select Git revision
  • better-specializations-for-realtuple
  • bugfix/fix_rigidbodymotion_difference
  • decasteljau
  • feature/ARRN-mod
  • feature/HM-numericalBenchmark
  • feature/HarmonicmapsBenchmark
  • feature/SimoFoxWithLocalFEfunctions
  • feature/bendingIsometries
  • feature/bendingIsometries-PBFE-Stiefel
  • feature/harmonicmapsAddons
  • feature/introduceRetractionNotion
  • feature/riemannianTRaddons
  • feature/simofoxBook
  • fix-fd-gradient-scaling
  • fix_localrodassembler_compiler_error
  • issue/vtk-namespace
  • make_rod-eoc_run
  • master default
  • releases/2.0-1
  • releases/2.1-1
20 results
Created with Raphaël 2.2.027Apr2322212019181716865Feb211Jan30Dec282529Oct24Jul232111Jun96519May1218Apr9Mar87643Jan31Dec20Nov1523Oct22171612428Sep5429Aug282724222120171514138321Feb58Jan54221Dec1413630Nov151324Aug2321Jun20151429May21Mar28Feb85Jan30Nov28155Sep23Aug18Julnew method to setup a solver using a truncated cg as the inner solver. This compiles, but it doesn't work yetremove several multigrid-specific data members, which were not even used anywaysuse an auto_ptr for the hesse Matrixmake the inner solver a pointer to a generic Solver class, in anticipation of using a truncated cg thereadapt to recent changes to the rod assembler and the riemannian tr solverbugfix: local stress must be a std::vectorremove historical cruft: don't create private copies of the local assemblermake localStiffness_ accessible from derived classeslocal assembler gets the material parameters nowmore involved boundary valuesnorm of the derivative in a dimension-independent waysimpler evaluate of the function value. Now only works with in the triangle, because there are some singularities lurking outside- More general initial valuesfd approximation of the gradient in a dimension-independent wayminor documentation fixtemporary: use the fd approximation of the gradient, and add some debugging codelast bugfix extended to all dimensionsBUGFIX: coordinates on subsimplices need to be normalized (only 2d, so far)bugfix: we need the derivative of the geodesic interpolation in the gradient computation (chain rule), not the derivative of the velocity interpolationfor debugging: add a method to compute gradients of local geodesic fe functions by finite-difference approximation (only 2d)bugfix: compute matrix occupation pattern on demandbugfix: method result wasn't returnedactually construct and call the solverimplement interpolation of the gradient for members of SO(3), some of this is still educated guesswork, thoughcomplete implementation, not tested beyond compilation yet, thoughexport 'EmbeddedTangentVector', the type of the tangentvector in the surrounding Euclidean spaceLocal interpolation for geodesic finite element functions centralized in one classcreate local assembler and hand it to the global oneaccept local assembler as constructor argumentinclude missing headersuse coordinate type from TargetSpaceexport type used for coordinatesexport type used for coordinatesmake binary writing of iterates work for all TargetSpace typesthe exponential map from a given point is moved from the solver into the manifold classesinclude missing headerTake correction vector blocksize from the target spacemake the target space a template parameteradd new method 'difference'RodSolver is now called RiemannianTrustRegionSolver, and can be parametrized with the TargetSpace. Some remains of RigidBodyMotion are still there, though.