Michael Mara

Sorry for the delay on this, I will be trying to upgrade my personal development machines to CUDA 10 this month, so hopefully can give you a real answer soon.

The binary we distribute with OpenMesh is compiled using VS2013, other versions (like 2015) may not work and you'll need to compile it yourself, like in #105.

Hello! Thanks for the feedback. I've cherrypicked your #include changes (except the explicit path to cuda_runtime.h) and pushed to master. I do not have a spare machine to test Ubuntu...

You could encode this using a graph, with edges between the image pixels and coefficients. This is an explicit encoding of W*H*K different edges; there isn't a natural implicit connectivity...

I think solving the problem with `insert_dummy_energies_for_unused_unknown_index_spaces` should be easiest by solving the general problem: Opt should support Unknowns in multiple domains, so long as the only interaction between those...

@Kalkasas What do you mean by a difference? From a solver where all the parameters are put in the same index space? When we fix the `TUnknownType.__apply` issue, the results...

I would not mind a parallel build system committed in the tree; but it is unlikely to be the main build system in the near future.

Not in the current version of Opt. We are working on a larger-scale revision to how Opt handle scheduling units of work that would likely subsume this concern. Out of...

That sounds like a memory leak to me! I'll work on tracking that down this weekend. As for a workaround, you could define some maximum problem size and have dummy...

That's consistent with the memory leak theory; I should have a fix by tomorrow for you to try.