libtorch

An open source machine learning framework that accelerates the path from research prototyping to production deployment.

Latest version: 2.5.0

Add to xmake.lua

add_requires("libtorch 2.5.0")

linux: i386, x86_64, armv7, armv7s, arm64, mips, mips64, mipsel, mips64el, loong64, riscv64, s390x, ppc, ppc64, ppc64el, sparc64
macosx: x86_64, arm64
windows: x64

blas: Set BLAS vendor. · Type: string · Default: openblas · Values: mkl, openblas, eigen
cuda: Enable CUDA support. · Type: boolean · Default: false
distributed: Enable distributed support. · Type: boolean · Default: false
ninja: Use ninja as build tool. · Type: boolean
openmp: Use OpenMP for parallel code. · Type: boolean · Default: true
protobuf-cpp: Use protobuf from xrepo. · Type: boolean · Default: false
pybind11: Use pybind11 from xrepo. · Type: boolean · Default: false
python: Build python interface. · Type: boolean · Default: false
shared: Build shared library. · Type: boolean · Default: true

Install the package independently of a project and inspect its available configuration.

xrepo install "libtorch 2.5.0"

xrepo info "libtorch"

Declare the requirement, then attach the library to the target that uses it.

add_requires("libtorch 2.5.0")

target("demo")
    set_kind("binary")
    add_files("src/*.cpp")
    add_packages("libtorch")

Open a shell with the environment exported by this package. Use the executable names documented by the package.

xrepo env -b "libtorch 2.5.0" shell