You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thank you for your information. My computer is only equipped with a 2070
GPU. I can eventually run the code by optimizing GPU resource usage.
Chenjunjie Wang
YihangChen ***@***.***> 于2024年6月8日周六 20:49写道:
Hey do you know how to deal with CUDA out of memory during rendering process?
encoding_param_num=1766272, size=0.2105560302734375MB. [08/06 19:57:10] Reading camera 251/251 [08/06 19:57:11] start fetching data from ply file [08/06 19:57:11] Loading Training Cameras [08/06 19:57:11] Loading Test Cameras [08/06 19:57:15] Initial voxel_size: 0.01 [08/06 19:57:16] Number of points at initialisation : 114293 [08/06 19:57:16] anchor_bound_updated [08/06 19:57:16] Training progress: 100%|███████████████████████████████████████████| 1000/1000 [01:00<00:00, 16.49it/s, Loss=0.0915876] 2024-06-08 19:58:17,128 - INFO: [ITER 1000] Saving Gaussians 2024-06-08 19:58:18,246 - INFO: Total Training time: 60.38757371902466 2024-06-08 19:58:18,351 - INFO: Training complete. 2024-06-08 19:58:18,351 - INFO: Starting Rendering~ hash_params: True 4 13 (18, 24, 33, 44, 59, 80, 108, 148, 201, 275, 376, 514) 15 (130, 258, 514, 1026) True False False [08/06 19:58:18] encoding_param_num=1766272, size=0.2105560302734375MB. [08/06 19:58:18] Loading trained model at iteration 1000 [08/06 19:58:18] Reading camera 251/251 [08/06 19:58:19] start fetching data from ply file [08/06 19:58:19] Loading Training Cameras [08/06 19:58:19] Loading Test Cameras [08/06 19:58:22] Rendering progress: 0%| | 0/32 [00:01<?, ?it/s] Traceback (most recent call last): File "train.py", line 669, in <module> visible_count = render_sets(args, lp.extract(args), -1, pp.extract(args), wandb=wandb, logger=logger, x_bound_min=x_bound_min, x_bound_max=x_bound_max) File "train.py", line 472, in render_sets t_test_list, visible_count = render_set(dataset.model_path, "test", scene.loaded_iter, scene.getTestCameras(), gaussians, pipeline, background) File "train.py", line 403, in render_set render_pkg = render(view, gaussians, pipeline, background, visible_mask=voxel_visible_mask) File "C:\Users\jay\Desktop\HAC\gaussian_renderer\__init__.py", line 225, in render cov3D_precomp = None) File "C:\Users\jay\AppData\Local\anaconda3\envs\HAC_env\lib\site-packages\torch\nn\modules\module.py", line 1102, in _call_impl return forward_call(*input, **kwargs) File "C:\Users\jay\AppData\Local\anaconda3\envs\HAC_env\lib\site-packages\diff_gaussian_rasterization\__init__.py", line 222, in forward raster_settings, File "C:\Users\jay\AppData\Local\anaconda3\envs\HAC_env\lib\site-packages\diff_gaussian_rasterization\__init__.py", line 41, in rasterize_gaussians raster_settings, File "C:\Users\jay\AppData\Local\anaconda3\envs\HAC_env\lib\site-packages\diff_gaussian_rasterization\__init__.py", line 92, in forward num_rendered, color, radii, geomBuffer, binningBuffer, imgBuffer = _C.rasterize_gaussians(*args) RuntimeError: CUDA out of memory. Tried to allocate 8.54 GiB (GPU 0; 8.00 GiB total capacity; 147.41 MiB already allocated; 5.60 GiB free; 264.00 MiB 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
The text was updated successfully, but these errors were encountered: