jiaw-z / CoR-GS

[ECCV'24] CoR-GS: Sparse-View 3D Gaussian Splatting via Co-Regularization
https://jiaw-z.github.io/CoR-GS/
Other
69 stars 5 forks source link

RuntimeError: tabulate: failed to synchronize: cudaErrorInvalidConfiguration: invalid configuration argument #25

Open 2927803072 opened 2 months ago

2927803072 commented 2 months ago

[500, 2000, 3000, 5000, 7000, 10000, 15000, 30000] Optimizing output/llff/fern Output folder: output/llff/fern [08/08 16:52:17] Tensorboard not available: not logging progress [08/08 16:52:17] args.source_path is data/nerf_llff_data/fern [08/08 16:52:17] ############ load llff ############ [08/08 16:52:17] Reading camera 20/20 [08/08 16:52:17] 2.8834194898605348 cameras_extent [08/08 16:52:17] Loading Training Cameras 1.0 [08/08 16:52:17] 3it [00:03, 1.24s/it] Loading Test Cameras 1.0 [08/08 16:52:21] 3it [00:01, 2.77it/s] Number of points at initialisation : 0 [08/08 16:52:27] Traceback (most recent call last): File "train.py", line 472, in training(lp.extract(args), op.extract(args), pp.extract(args), args) File "train.py", line 66, in training scene = Scene(args, gaussians, shuffle=False) File "/home/jy/demo/CoR-GS-main/scene/init.py", line 120, in init self.gaussians.create_from_pcd(scene_info.point_cloud, self.cameras_extent) File "/home/jy/demo/CoR-GS-main/scene/gaussian_model.py", line 144, in create_from_pcd dist2 = torch.clamp_min(distCUDA2(fused_point_cloud)[0], 0.0000001) RuntimeError: tabulate: failed to synchronize: cudaErrorInvalidConfiguration: invalid configuration argument

Hello, I have encountered a certain issue for which I am uncertain of the cause. Could you please provide guidance on how to resolve it? Your assistance would be greatly appreciated.

2927803072 commented 2 months ago

I use 3090, ubuntu 20.04,cuda11.8