search

qiqihaer / RandLA-Net-pytorch

RandLA-Net's implementation with Pytorch
MIT License
127 stars 20 forks source link

About pytorch version. #8

Open a18700 opened 3 years ago

a18700 commented 3 years ago

Hello,

I got errors below, when to experiment with your codes.

I think this is a pytorch version issue, thereby i want to know the implementation settings.

/workspace/gt_cvpr/gt_codes/NewBaseLine/RandLA-Net-pytorch/RandLANet.py:264: UserWarning: torch.range is deprecated in favor of torch.arange and will be removed in 0.5. Note that arange generates values in [start; end), not [start; end]. reducing_list = torch.range(0, cfg.num_classes).long().cuda() Traceback (most recent call last): File "main_SemanticKITTI.py", line 209, in train(start_epoch) File "main_SemanticKITTI.py", line 189, in train train_one_epoch() File "main_SemanticKITTI.py", line 110, in train_one_epoch loss.backward() File "/opt/conda/lib/python3.7/site-packages/torch/tensor.py", line 195, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph) File "/opt/conda/lib/python3.7/site-packages/torch/autograd/init.py", line 99, in backward allow_unreachable=True) # allow_unreachable flag RuntimeError: cuDNN error: CUDNN_STATUS_NOT_SUPPORTED. This error may appear if you passed in a non-contiguous input.

qiqihaer commented 3 years ago

Hello,

I got errors below, when to experiment with your codes.

I think this is a pytorch version issue, thereby i want to know the implementation settings.

/workspace/gt_cvpr/gt_codes/NewBaseLine/RandLA-Net-pytorch/RandLANet.py:264: UserWarning: torch.range is deprecated in favor of torch.arange and will be removed in 0.5. Note that arange generates values in [start; end), not [start; end]. reducing_list = torch.range(0, cfg.num_classes).long().cuda() Traceback (most recent call last): File "main_SemanticKITTI.py", line 209, in train(start_epoch) File "main_SemanticKITTI.py", line 189, in train train_one_epoch() File "main_SemanticKITTI.py", line 110, in train_one_epoch loss.backward() File "/opt/conda/lib/python3.7/site-packages/torch/tensor.py", line 195, in backward torch.autograd.backward(self, gradient, retain_graph, create_graph) File "/opt/conda/lib/python3.7/site-packages/torch/autograd/init.py", line 99, in backward allow_unreachable=True) # allow_unreachable flag RuntimeError: cuDNN error: CUDNN_STATUS_NOT_SUPPORTED. This error may appear if you passed in a non-contiguous input.

CUDA9.0, Pytorch1.1