search

dmsm / HodgeNet

Code for HodgeNet: Learning Spectral Geometry on Triangle Meshes, in SIGGRAPH 2021.
MIT License
79 stars 9 forks source link

code for prediction and visualization on segmentation output #2

Open amiltonwong opened 3 years ago

amiltonwong commented 3 years ago

Hi, @dmsm,

Could you also provide the code for prediction and visualization on segmentation output?

Thanks~

Dengzhi-USTC commented 3 years ago

Hi, @dmsm @amiltonwong , Can you retrain this code? I have been waiting in the line 78 of the hodgeautograd.py( ctxeigenvectors.copy(torch.from_numpy(reshaped_eigenvectors).to(device))) when run on the cpu; Did you meet this issure?

Thanks

Dengzhi-USTC commented 3 years ago

Hi, @dmsm @amiltonwong , Can you retrain this code? I have been waiting in the line 78 of the hodgeautograd.py( ctxeigenvectors.copy(torch.from_numpy(reshaped_eigenvectors).to(device))) when run on the cpu; Did you meet this issure?

Thanks

I solved this problem, need considerate the version of the torch.multiprocess. (Add the torch.multiprocessing.set_start_method('spawn', force=True) is ok)

ManjitKuma commented 3 months ago

Have anyone tried to test and visualize the output of the model? I tried but not getting expected result .if anyone got then please give me some suggestion ..