Introduction

This document describes how to build the Paddle Inference C++ GPU inference library from the PaddlePaddle source code on openEuler and verify the installation.

Paddle Inference is a C++ inference library deployed on the server by PaddlePaddle. It provides C++ APIs, header files, dynamic libraries, and static libraries. The official open-source project of Paddle Inference does not provide a precompiled binary installation package for openEuler AArch64. You are advised to use the source code build product as the delivery and installation object.