jackhu-bme
commented
1 year ago you should install a higher version of CUDA and pytorch,as your gpu card is pretty good. At least pytorch 1.11.1 + CUDA11.3, I guess. This works for my 3090 card, which is also above sm75 arch.
NVIDIA GeForce RTX 3080 Ti with CUDA capability sm_86 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70 sm_75. If you want to use the NVIDIA GeForce RTX 3080 Ti GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/ warnings.warn(incompatible_device_warn.format(device_name, capability, " ".join(arch_list), device_name)) File "train.py", line 357, in
main()
File "train.py", line 316, in main
train_log = train(config, train_loader, model, criterion, optimizer)
File "train.py", line 124, in train
output = model(input)
File "/home/public_2/develop/Anaconda3/envs/unext/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(*input, *kwargs)
File "/home/public_2/qht/UNeXt-pytorch/archs.py", line 279, in forward
out = F.relu(F.max_pool2d(self.ebn1(self.encoder1(x)),2,2))
File "/home/public_2/develop/Anaconda3/envs/unext/lib/python3.6/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
result = self.forward(input, **kwargs)
File "/home/public_2/develop/Anaconda3/envs/unext/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 423, in forward
return self._conv_forward(input, self.weight)
File "/home/public_2/develop/Anaconda3/envs/unext/lib/python3.6/site-packages/torch/nn/modules/conv.py", line 420, in _conv_forward
self.padding, self.dilation, self.groups)
RuntimeError: CUDA error: no kernel image is available for execution on the device