RuntimeError: Failed to set maximum shared memory size (requested 147456 bytes), try lowering tile_size.

[lingzhenfei]

Traceback (most recent call last):
File "train.py", line 618, in
training(lp, op, pp, exp_name, args.test_iterations, args.save_iterations, args.checkpoint_iterations, args.start_checkpoint, wandb, logger)
File "train.py", line 249, in training
total_loss.backward()
File "/storage/lzf_home/anaconda3/envs/objectgs/lib/python3.8/site-packages/torch/_tensor.py", line 492, in backward
torch.autograd.backward(
File "/storage/lzf_home/anaconda3/envs/objectgs/lib/python3.8/site-packages/torch/autograd/init.py", line 251, in backward
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
File "/storage/lzf_home/anaconda3/envs/objectgs/lib/python3.8/site-packages/torch/autograd/function.py", line 288, in apply
return user_fn(self, *args)
File "/storage/lzf_home/3dgs/ObjectGS/gsplat-object/gsplat/cuda/_wrapper.py", line 1944, in backward
) = _make_lazy_cuda_func("rasterize_to_pixels_bwd_2dgs")(
File "/storage/lzf_home/3dgs/ObjectGS/gsplat-object/gsplat/cuda/_wrapper.py", line 14, in call_cuda
return getattr(_C, name)(*args, **kwargs)
RuntimeError: Failed to set maximum shared memory size (requested 147456 bytes), try lowering tile_size.

[RuijieZhu94]

Hi @lingzhenfei, TBH, I never met this issue. Maybe your GPU memories is not enough? You may ask GPT for the help:)

[RuijieZhu94]

Hi @lingzhenfei, do you fix it now?