From 2ee5c95f9e67a2d6489683f461cc581c32e955de Mon Sep 17 00:00:00 2001 From: Massimiliano Bonomi Date: Fri, 29 Sep 2023 19:13:23 +0200 Subject: [PATCH] add example to download CUDA-enabled LibTorch 2.0.0 --- user-doc/Installation.md | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/user-doc/Installation.md b/user-doc/Installation.md index e95b3447af..8ce75f21e3 100644 --- a/user-doc/Installation.md +++ b/user-doc/Installation.md @@ -320,7 +320,14 @@ wget https://download.pytorch.org/libtorch/cpu/libtorch-cxx11-abi-shared-with-de unzip libtorch-cxx11-abi-shared-with-deps-2.0.0+cpu.zip ; \endverbatim -The location of the include and library files need to be exported in the environment: +If you have a GPU, you might want to use the CUDA-accelerated version of LibTorch. For example, the following script downloads the `libtorch-shared-with-deps-2.0.0%2Bcu117.zip` (2.0.0, GPU, Cuda 11.7, pre-cxx11 ABI binary). + +\verbatim +wget https://download.pytorch.org/libtorch/cu117/libtorch-shared-with-deps-2.0.0%2Bcu117.zip +unzip libtorch-shared-with-deps-2.0.0+cu117.zip +\endverbatim + +In both CPU and GPU cases, the location of the include and library files need to be exported in the environment: \verbatim LIBTORCH=${PWD}/libtorch