Skip to content

Commit

Permalink
Merge pull request #1735 from TeslaZhao/develop
Browse files Browse the repository at this point in the history
Update doc
  • Loading branch information
TeslaZhao authored Apr 18, 2022
2 parents aabecfa + 02d581b commit ee8b5e2
Show file tree
Hide file tree
Showing 22 changed files with 828 additions and 131 deletions.
11 changes: 8 additions & 3 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -40,13 +40,20 @@ The goal of Paddle Serving is to provide high-performance, flexible and easy-to-
- Support service monitoring, provide prometheus-based performance statistics and port access


<h2 align="center">Tutorial</h2>
<h2 align="center">Tutorial and Papers</h2>


- AIStudio tutorial(Chinese) : [Paddle Serving服务化部署框架](https://www.paddlepaddle.org.cn/tutorials/projectdetail/2538249)
- AIStudio OCR practice(Chinese) : [基于PaddleServing的OCR服务化部署实战](https://aistudio.baidu.com/aistudio/projectdetail/3630726)
- Video tutorial(Chinese) : [深度学习服务化部署-以互联网应用为例](https://aistudio.baidu.com/aistudio/course/introduce/19084)
- Edge AI solution(Chinese) : [基于Paddle Serving&百度智能边缘BIE的边缘AI解决方案](https://mp.weixin.qq.com/s/j0EVlQXaZ7qmoz9Fv96Yrw)

- Paper : [JiZhi: A Fast and Cost-Effective Model-As-A-Service System for
Web-Scale Online Inference at Baidu](https://arxiv.org/pdf/2106.01674.pdf)
- Paper : [ERNIE 3.0 TITAN: EXPLORING LARGER-SCALE KNOWLEDGE
ENHANCED PRE-TRAINING FOR LANGUAGE UNDERSTANDING
AND GENERATION](https://arxiv.org/pdf/2112.12731.pdf)

<p align="center">
<img src="doc/images/demo.gif" width="700">
</p>
Expand Down Expand Up @@ -90,8 +97,6 @@ The first step is to call the model save interface to generate a model parameter
- [Analyze and optimize performance](doc/Python_Pipeline/Performance_Tuning_EN.md)
- [TensorRT dynamic Shape](doc/TensorRT_Dynamic_Shape_EN.md)
- [Benchmark(Chinese)](doc/Python_Pipeline/Benchmark_CN.md)
- Our Paper: [JiZhi: A Fast and Cost-Effective Model-As-A-Service System for
Web-Scale Online Inference at Baidu](https://arxiv.org/pdf/2106.01674.pdf)
- Client SDK
- [Python SDK(Chinese)](doc/C++_Serving/Introduction_CN.md#42-多语言多协议Client)
- [JAVA SDK](doc/Java_SDK_EN.md)
Expand Down
17 changes: 12 additions & 5 deletions README_CN.md
Original file line number Diff line number Diff line change
Expand Up @@ -39,11 +39,18 @@ Paddle Serving依托深度学习框架PaddlePaddle旨在帮助深度学习开发
- 支持服务监控,提供基于普罗米修斯的性能数据统计及端口访问


<h2 align="center">教程</h2>

- AIStudio教程-[Paddle Serving服务化部署框架](https://www.paddlepaddle.org.cn/tutorials/projectdetail/2538249)
- 视频教程-[深度学习服务化部署-以互联网应用为例](https://aistudio.baidu.com/aistudio/course/introduce/19084)
- 边缘AI解决方案-[基于Paddle Serving&百度智能边缘BIE的边缘AI解决方案](https://mp.weixin.qq.com/s/j0EVlQXaZ7qmoz9Fv96Yrw)
<h2 align="center">教程与论文</h2>

- AIStudio 使用教程 : [Paddle Serving服务化部署框架](https://www.paddlepaddle.org.cn/tutorials/projectdetail/2538249)
- AIStudio OCR实战 : [基于PaddleServing的OCR服务化部署实战](https://aistudio.baidu.com/aistudio/projectdetail/3630726)
- 视频教程 : [深度学习服务化部署-以互联网应用为例](https://aistudio.baidu.com/aistudio/course/introduce/19084)
- 边缘AI 解决方案 : [基于Paddle Serving&百度智能边缘BIE的边缘AI解决方案](https://mp.weixin.qq.com/s/j0EVlQXaZ7qmoz9Fv96Yrw)

- 论文 : [JiZhi: A Fast and Cost-Effective Model-As-A-Service System for
Web-Scale Online Inference at Baidu](https://arxiv.org/pdf/2106.01674.pdf)
- 论文 : [ERNIE 3.0 TITAN: EXPLORING LARGER-SCALE KNOWLEDGE
ENHANCED PRE-TRAINING FOR LANGUAGE UNDERSTANDING
AND GENERATION](https://arxiv.org/pdf/2112.12731.pdf)

<p align="center">
<img src="doc/images/demo.gif" width="700">
Expand Down
10 changes: 9 additions & 1 deletion doc/Offical_Docs/3-0_QuickStart_Int_CN.md
Original file line number Diff line number Diff line change
@@ -1,2 +1,10 @@
# 部署案例概述
# 快速开始案例部署

您可以通过以下 Paddle Serving 快速开始案例,分别了解到 C++ Serving 与 Python Pipeline 2种框架的部署方法。
- [使用 C++ Serving 部署 Resnet50 模型案例]()
- [使用 Python Pipeline 部署 OCR 模型案例]()

通过阅读以下内容掌握 Paddle Serving 基础功能以及2种框架特性和使用指南:
- [基础功能]()
- [进阶 C++ Serving 介绍]()
- [进阶 Python Pipeline 介绍]()
12 changes: 12 additions & 0 deletions doc/Offical_Docs/7-0_Python_Pipeline_Int_CN.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
# Python Pipeline 框架

在许多深度学习框架中,模型服务化部署通常用于单模型的一键部署。但在 AI 工业大生产的背景下,端到端的单一深度学习模型不能解决复杂问题,多个深度学习模型组合使用是解决现实复杂问题的常规手段,如文字识别 OCR 服务至少需要检测和识别2种模型;视频理解服务一般需要视频抽帧、切词、音频处理、分类等多种模型组合实现。当前,通用多模型组合服务的设计和实现是非常复杂的,既要能实现复杂的模型拓扑关系,又要保证服务的高并发、高可用和易于开发和维护等。

Paddle Serving 实现了一套通用的多模型组合服务编程框架 Python Pipeline,不仅解决上述痛点,同时还能大幅提高 GPU 利用率,并易于开发和维护。

Python Pipeline 使用案例请阅读[Python Pipeline 快速部署案例](./3-2_QuickStart_Pipeline_OCR_CN.md)

通过阅读以下内容掌握 Python Pipeline 设计方案、高阶用法和优化指南等。
- [Python Pipeline 框架设计](7-1_Python_Pipeline_Design_CN.md)
- [Python Pipeline 高阶用法](7-2_Python_Pipeline_Senior_CN.md)
- [Python Pipeline 优化指南](7-3_Python_Pipeline_Optimize_CN.md)
Loading

0 comments on commit ee8b5e2

Please sign in to comment.