pytorch on nano jetson

Hi,
i have tried to install pytorch 1.3.0 for python 3.6 according to your instructions in https://devtalk.nvidia.com/default/topic/1049071/pytorch-for-jetson-nano-version-1-3-0-now-available/

after that i followed your instructions for installing torchvision with the proper branch v0.5.0.

when trying to import torchvision i get the following error:
“RuntimeError:
builtin cannot be used as a value:
at /usr/local/lib/python3.6/dist-packages/torchvision-0.5.0a0+07cbb46-py3.6-linux-aarch64.egg/torchvision/models/detection/_utils.py:14:56
def zeros_like(tensor, dtype):
# type: (Tensor, int) → Tensor
return torch.zeros_like(tensor, dtype=dtype, layout=tensor.layout,
~~~~~~~~~~~~~ <— HERE
device=tensor.device, pin_memory=tensor.is_pinned())”

is there a way to solve it?

Hi NNresearch, please refer to this post regarding the workaround (install torchvision 0.4.2)

Looks like it’s a known issue in PyTorch and is slated to be fixed in PyTorch 1.4 release.

Hi,
thanks for the reply , it solved the issue.