hi, thanks for your great work, some errors when train with custom datasets
Loading cameras: 3500 for training and 500 for testing Number of points at initialisation : 1269404 Training progress: 0%| | 0/40000 [00:00<?, ?it/s]Traceback (most recent call last): File "gaussian_surfels/train.py", line 300, in <module> training(lp.extract(args), op.extract(args), pp.extract(args), args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from) File "gaussian_surfels/train.py", line 145, in training loss.backward() File "conda_envs/envs/surfel_gs/lib/python3.10/site-packages/torch/_tensor.py", line 396, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "conda_envs/envs/surfel_gs/lib/python3.10/site-packages/torch/autograd/__init__.py", line 173, in backward Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass RuntimeError: Function _RasterizeGaussiansBackward returned an invalid gradient at index 2 - got [0, 0, 3] but expected shape compatible with [0, 16, 3]
@turandai Looking forward to your reply, thanks.
turandai
commented
6 months ago 
cv-lab-x
hi, thanks for your great work, some errors when train with custom datasets
Loading cameras: 3500 for training and 500 for testing Number of points at initialisation : 1269404 Training progress: 0%| | 0/40000 [00:00<?, ?it/s]Traceback (most recent call last): File "gaussian_surfels/train.py", line 300, in <module> training(lp.extract(args), op.extract(args), pp.extract(args), args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, args.debug_from) File "gaussian_surfels/train.py", line 145, in training loss.backward() File "conda_envs/envs/surfel_gs/lib/python3.10/site-packages/torch/_tensor.py", line 396, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph, inputs=inputs) File "conda_envs/envs/surfel_gs/lib/python3.10/site-packages/torch/autograd/__init__.py", line 173, in backward Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass RuntimeError: Function _RasterizeGaussiansBackward returned an invalid gradient at index 2 - got [0, 0, 3] but expected shape compatible with [0, 16, 3]@turandai Looking forward to your reply, thanks.