Pytorch 1.3.1 binaries only support GPUs with CUDA capabilty>=3.7 as clarified by @ptrblck here
When using NMS from torchvision.ops on K80 (3,7), I’m getting this error:
CUDA error: no kernel image is available for execution on the device (nms_cuda at /tmp/pip-req-build-9d9zypi6/torchvision/csrc/cuda/nms_cuda.cu:127)
frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0x6d (0x7f096ecbce7d in /opt/anaconda3/lib/python3.7/site-packages/torch/lib/libc10.so)
frame #1: nms_cuda(at::Tensor const&, at::Tensor const&, float) + 0x8d1 (0x7f0918a32ece in /opt/anaconda3/lib/python3.7/site-packages/torchvision/_C.so)
frame #2: nms(at::Tensor const&, at::Tensor const&, float) + 0x183 (0x7f09189f6ed7 in /opt/anaconda3/lib/python3.7/site-packages/torchvision/_C.so)
Is this a bug/unintended behaviour because it should support it ? Am I missing something ?