RuntimeError: Cuda error: 2[cudaMalloc(&m_gpuPtr, bytes);]`

[ethanshenze]

When I used the pictures taken by realsense d435i to test, I encountered such an error. I couldn't find the reason. What is the cause of this problem? (I just replaced the data in the mustard0 folder with my own data)

error like this: [register()] Welcome Module Utils load on device 'cuda:0' took 144.14 ms [register()] poses:(252, 4, 4) [register()] after viewpoint, add_errs min:-1.0 /opt/conda/envs/my/lib/python3.8/site-packages/torch/__init__.py:614: UserWarning: torch.set_default_tensor_type() is deprecated as of PyTorch 2.1, please use torch.set_default_dtype() and torch.set_default_device() as alternatives. (Triggered internally at ../torch/csrc/tensor/python_tensor.cpp:451.) _C._set_default_tensor_type(t) [predict()] ob_in_cams:(252, 4, 4) [predict()] self.cfg.use_normal:False [predict()] trans_normalizer:[0.019999999552965164, 0.019999999552965164, 0.05000000074505806], rot_normalizer:0.3490658503988659 [predict()] making cropped data [make_crop_data_batch()] Welcome make_crop_data_batch [make_crop_data_batch()] make tf_to_crops done RuntimeError: Cuda error: 2[cudaMalloc(&m_gpuPtr, bytes);]

[wenbowen123]

https://github.com/NVlabs/FoundationPose/issues/25#issuecomment-2037719050 can you try the suggestions there?

It's likely your data is not formatted correctly.