search

wenbin-lin / OcclusionFusion

OcclusionFusion: realtime dynamic 3D reconstruction based on single-view RGB-D
https://wenbin-lin.github.io/OcclusionFusion/
256 stars 41 forks source link

How to decrease memory usage? #12

Open luxious opened 2 years ago

luxious commented 2 years ago

This is a really great project, we try to run the demo in our PC (NVIDIA GeForce RTX 3080 Ti). But we encountered this error, can anyone tell me how to resolve this issue. we have tried setting max_split_size_mb but not worked.

File "E:\projects\OcclusionFusion\model.py", line 89, in forward
  feature7 = self.layer72(feature7, edge_indexes[0])
...
RuntimeError: CUDA out of memory. Tried to allocate 2.00 MiB (GPU 0; 12.00 GiB total capacity; 11.10 GiB already allocated; 0 bytes free; 11.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation.  See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
Barbany commented 1 year ago

I created a PR for that. You can check it at https://github.com/wenbin-lin/OcclusionFusion/pull/14