apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-31 03:46:40 +08:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity
Files
85e1c647f63eade56c947762e4adb37ebc0b2f18
FastDeploy/serving/docs/zh_CN/compile.md
heliqi 831ef42207 fix paddle-mkldnn bug (#366) 
* fix input name bug

* fix paddle-mkldnn bug
2022-10-13 21:37:31 +08:00

970 B
Raw Blame History

服务化部署镜像编译

本文档介绍如何制作FastDploy镜像

制作GPU镜像

FastDploy发布的GPU镜像基于Triton Inference Server的21.10版本进行制作，如果有其他CUDA版本需求，可以参照NVIDIA 官网中展示的版本信息修改Dockerfile和scripts中的脚本.

# 进入serving目录执行脚本编译fastdeply和服务化的backend
cd serving
bash scripts/build.sh

# 退出到FastDeploy主目录，制作镜像
cd ../
docker build -t paddlepaddle/fastdeploy:0.3.0-gpu-cuda11.4-trt8.4-21.10 -f serving/Dockerfile .

制作CPU镜像

# 进入serving目录执行脚本编译fastdeply和服务化的backend
cd serving
bash scripts/build.sh OFF

# 退出到FastDeploy主目录，制作镜像
cd ../
docker build -t paddlepaddle/fastdeploy:0.3.0-cpu-only -f serving/Dockerfile_cpu .