GordonWei

> the pytorch model I have trianed for 5days with gtx2080ti makes an unreasonable result.hope someone can share > a good pretrained model. Thank you for this information. I have...

Same question. There is no epoch in args.

> Excuse my poor understanding towards EMD. I test geomloss with the following code: > > ``` > points1 = torch.rand(2, 1024, 2, requires_grad=True).cuda() > points2 = torch.rand(2, 1024, 2).cuda()...

Hey, I've met just the same problem with you. I tried to train a 2 stacks HG network with the original code and params. I guess this is cause by...

@djangogo for me, setting a smaller learning rate could help, for example, you may set it to be around 1e-5, other tricks on adjusting learning rate could also be helpful.

Looking forward to the support of interaction with the scene. I think it will make Polyscope suitable in more applications.

> The current master of geometry-central fails to build for me on: > > ``` > CMake Error at deps/geometry-central/deps/CMakeLists.txt:158 (file): > file failed to create symbolic link > 'C:/Users/ernes/Downloads/V-Sekai.intrinsic-triangulations/build/deps/geometry-central/deps/include/nanoflann/nanoflann.hpp':...

Similar problem. The generator loss go downs from a thousand to around 50, but the discriminator loss keep jittering around some values. Anyone gets an idea on how to solve...

> Are you trying to run this over SSH? Hi, what if I want to use Meshrender on a server without any display?

@mmatl Thanks!