how run camm on one RTX 3090 GPU?

[limaolin2017]

Hi, I encounter an error:

Traceback (most recent call last):
File "/gpfs/home/mli/camm/main.py", line 40, in
app.run(main)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/absl/app.py", line 312, in run
_run_main(main, args)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/absl/app.py", line 258, in _run_main
sys.exit(main(argv))
File "/gpfs/home/mli/camm/main.py", line 37, in main
trainer.train()
File "/gpfs/home/mli/camm/nnutils/train_utils.py", line 904, in train
self.train_one_epoch(epoch, log)
File "/gpfs/home/mli/camm/nnutils/train_utils.py", line 1129, in train_one_epoch
total_loss,aux_out = self.model(batch)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/nn/parallel/distributed.py", line 886, in forward
output = self.module(*inputs[0], **kwargs[0])
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/gpfs/home/mli/camm/nnutils/camm.py", line 686, in forward_default
mesh_rest = pytorch3d.structures.meshes.Meshes(
File "/gpfs/home/mli/camm/third_party/pytorch3d/pytorch3d/structures/meshes.py", line 406, in init
if len(self._num_faces_per_mesh.unique()) == 1:
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/_tensor.py", line 530, in unique
return torch.unique(self, sorted=sorted, return_inverse=return_inverse, return_counts=return_counts, dim=dim)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/_jit_internal.py", line 422, in fn
return if_false(*args, **kwargs)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/_jit_internal.py", line 422, in fn
return if_false(*args, **kwargs)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/functional.py", line 821, in _return_output
output, _, _ = _unique_impl(input, sorted, return_inverse, return_counts, dim)
File "/home/mli/.conda/envs/camm/lib/python3.9/site-packages/torch/functional.py", line 735, in _unique_impl
output, inverse_indices, counts = torch._unique2(
RuntimeError: std::bad_alloc

hardware device: one RTX 3090 GPU

[kts707]

You can try to see if the steps mentioned here would resolve the issue.

[limaolin2017]

Thank you for your reply! Its fixed by pip install all dependencies again.