nvcc fatal : Unsupported gpu architecture 'compute_75'

[wytcsuch]

when I run :python setup.py build develop
It shows：

/usr/local/cuda/bin/nvcc -DWITH_CUDA -I/home/yantianwang/detectron2/detectron2/layers/csrc -I/home/yantianwang/anaconda2/envs/pytorch/lib/python3.6/site-packages/torch/include -I/home/yantianwang/anaconda2/envs/pytorch/lib/python3.6/site-packages/torch/include/torch/csrc/api/include -I/home/yantianwang/anaconda2/envs/pytorch/lib/python3.6/site-packages/torch/include/TH -I/home/yantianwang/anaconda2/envs/pytorch/lib/python3.6/site-packages/torch/include/THC -I/usr/local/cuda/include -I/home/yantianwang/anaconda2/envs/pytorch/include/python3.6m -c /home/yantianwang/detectron2/detectron2/layers/csrc/ROIAlign/ROIAlign_cuda.cu -o build/temp.linux-x86_64-3.6/home/yantianwang/detectron2/detectron2/layers/csrc/ROIAlign/ROIAlign_cuda.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -DCUDA_HAS_FP16=1 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ -DTORCH_API_INCLUDE_EXTENSION_H -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_75,code=sm_75 -std=c++11 nvcc fatal : Unsupported gpu architecture 'compute_75' error: command '/usr/local/cuda/bin/nvcc' failed with exit status 1

Environment

GPU：GTX 2080TI
CUDA 9.2
cudnn 7.6.3
pytorch 1.13.0

[ppwwyyxx]

Please include details about the problem following the issue template.

[sophia-hxw]

I have the same problem and U can look the solution in the blog

    https://blog.csdn.net/sophia_xw/article/details/102561410

May be it's because U have more than one CUDA in your machine.
You can use 'mvcc -V' or 'cat /usr/local/cuda/version.txt' or 'ls /usr/bin -la' or 'stat Cuda' to see the version of your CUDA.

The solution is, build path and ld_library in ~/.bashrc

sudo gedit ~/.bashrc
export PATH=/usr/local/cuda-10.1/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-10.1/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
source ~/.bashrc

[freesunshine]

@wytcsuch You need CUDA 10.x for RTX 2080ti.

[jrojoCI3]

Thank you @sophia-hxw that was indeed the problem. Once added the 10.1 cuda version to the path (and the libraries) the compilation didn't complain.

[gmt710]

torch/torch7#1190