Docs for Installing C++ Distributions of PyTorch
needs update
#60372
Labels
module: docs
Related to our documentation, both in docs/ and docblocks
triaged
This issue has been looked at a team member, and triaged and prioritized into an appropriate module
馃摎 Documentation
In the C++ docs, it writes:
However, the
wget
links to a GPU version libtorch. I followed this tutorial on Ubuntu 20.04.2 LTS (GNU/Linux 5.4.0-73-generic x86_64), when I runcmake -DCMAKE_PREFIX_PATH=/absolute/path/to/libtorch ..
, I got a CMAKE errorImported target "torch" includes non-existent path "/pytorch/torch/lib"
:I checked some issues, and they said it's because the CUDA were not installed correctly. But I run it on the cheapest virtual server on Digital Ocean, and there's no GPU... Could it be that I downloaded the wrong one? But I was just following the docs...
Finally I deleted this version libtorch and downloaded a cpu version from https://pytorch.org/get-started/locally/ . Then everything went well as expected.
So I think shall we update the docs?
cc @brianjo @mruberry
The text was updated successfully, but these errors were encountered: