search

cvlab-kaist / RAIN-GS

Code for "Relaxing Accurate Initialization Constraint for 3D Gaussian Splatting" by Jaewoo Jung, Jisang Han, Honggyu An, Jiwon Kang, Seonghoon Park, and Seungryong Kim
https://ku-cvlab.github.io/RAIN-GS
MIT License
298 stars 16 forks source link

RuntimeError: Function _RasterizeGaussiansBackward returned an invalid gradient at index 2 - got [0, 0, 3] but expected shape compatible with [0, 16, 3] #17

Closed jiangyijin closed 1 month ago

jiangyijin commented 1 month ago

I encountered a problem; how can I solve it? Traceback (most recent call last): File "/data/RAIN-GS/train.py", line 295, in 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, args.dict) File "/data/RAIN-GS/train.py", line 115, in training loss.backward() File "/home/ubuntu/anaconda3/envs/3dgs/lib/python3.9/site-packages/torch/_tensor.py", line 487, in backward torch.autograd.backward( File "/home/ubuntu/anaconda3/envs/3dgs/lib/python3.9/site-packages/torch/autograd/init.py", line 200, 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]

crepejung00 commented 1 month ago

Hi, We currently found that there are some minor code errors in our current version, and we are refactoring the code. Sorry for the inconvenience and we will note you when the codes are fixed.