search

nv-tlabs / NKSR

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction
https://research.nvidia.com/labs/toronto-ai/NKSR
Other
765 stars 44 forks source link

RuntimeError: Not compiled with CUDA support #21

Closed tqlhq closed 1 year ago

tqlhq commented 1 year ago

When I run this code:recons_simple.py,it threats an error:

Traceback (most recent call last): File "/home/vector4d/NKSR/examples/recons_simple.py", line 26, in field = reconstructor.reconstruct(input_xyz, input_normal, detail_level=1.0) File "/home/vector4d/anaconda3/envs/nksr2/lib/python3.10/site-packages/nksr/init.py", line 255, in reconstruct feat = self.network.encoder(xyz, feat, svh, 0) File "/home/vector4d/anaconda3/envs/nksr2/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1501, in _call_impl return forward_call(*args, *kwargs) File "/home/vector4d/anaconda3/envs/nksr2/lib/python3.10/site-packages/nksr/nn/encdec.py", line 105, in forward pooled = scatter_max(pts_feature, vid, dim=0, dim_size=grid.num_voxels)[0] File "/home/vector4d/anaconda3/envs/nksr2/lib/python3.10/site-packages/torch_scatter/scatter.py", line 72, in scatter_max return torch.ops.torch_scatter.scatter_max(src, index, dim, out, dim_size) File "/home/vector4d/anaconda3/envs/nksr2/lib/python3.10/site-packages/torch/_ops.py", line 502, in call return self._op(args, **kwargs or {}) RuntimeError: Not compiled with CUDA support

My operating environment is python 3.10 torch2.0.0+cu118 (in conda) How to solve this issue?Please help me !!!

heiwang1997 commented 1 year ago

This seems to be a torch_scatter's problem. Did you create the conda environment using environment.yml? Or did you compile torch_scatter yourself?

tqlhq commented 1 year ago

Thank you very much for your answer. I have already solved this problem. Thank you again

---Original--- From: "Jiahui @.> Date: Mon, Jul 3, 2023 19:48 PM To: @.>; Cc: @.**@.>; Subject: Re: [nv-tlabs/NKSR] RuntimeError: Not compiled with CUDA support(Issue #21)

This seems to be a torch_scatter's problem. Did you create the conda environment using environment.yml? Or did you compile torch_scatter yourself?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

heiwang1997 commented 1 year ago

OK it's great to hear that! Happy to help again.