NOTE:
These packages are not open source and do not contain any source code. Instead, in order to "build" the exported target(s) they download (potentially large) pre-built binaries provided by NVIDIA for the target platform.CI for these packages are disabled due to the above.
Supported platforms/compilers are Windows/MSVC and Linux/GCC or Clang (x86_64).
This is a build2 package repository for NVIDIA® TensorRT™ SDK.
NVIDIA TensorRT is an SDK for high-performance deep learning inference, includes a deep learning inference optimizer and runtime that delivers low latency and high throughput for inference applications.
This file contains setup instructions and other details that are more
appropriate for package development rather than consumption. If you want to use
nvidia-tensorrt in your build2-based project, then instead see the accompanying
package(s) corresponding README.md file.
The development setup for nvidia-tensorrt uses the standard bdep-based workflow.
For example:
git clone https://github.com/build2-packaging/nvidia-tensorrt
cd nvidia-tensorrt
bdep init -C @msvc cc config.cxx=cl.exe
bdep update
bdep test
Or on Linux:
bdep init -C @gcc cc config.cxx=g++
bdep update
bdep test