中文 | English Readme | deepwiki

<h1 align="center"> CGraph 说明文档 </h1> <img align="right" src="https://github.com/ChunelFeng/CGraph/blob/main/doc/image/CGraph%20Author.jpg" width="256px">

CGraph is a cross-platform Directed Acyclic Graph framework based on pure C++ without any 3rd-party dependencies. You, with it, can build your own operators simply, and describe any running schedules as you need, such as dependence, parallelling, aggregation, conditional and so on. Python APIs are also supported to build your pipeline. Tutorials and contact information are shown as follows. Please get in touch with us for free if you need more about this repository.

一. 简介

CGraph中文名为【色丶图】，是一套无任何第三方依赖的跨平台图流程执行框架。通过GPipeline(流水线)底层调度，提供了包含依赖元素依次执行、非依赖元素并发执行，支持暂停、恢复、超时设定的 eDAG 调度功能。

使用者只需继承GNode(节点)类，实现子类的run()方法，并根据需要设定依赖关系，即可实现任务的图化执行或流水线执行。还可以通过设定各种包含多节点信息的GGroup(组)，自行控制图的条件判断、循环和并发执行逻辑。

CGraph Skeleton

本工程使用纯C++11标准库编写，无任何第三方依赖，兼容MacOS、Linux、Windows和Android系统。支持本地编译和二次开发，并且提供Python版本：pycgraph。编译和安装方法，请参考 CGraph 编译说明

详细功能介绍和用法，请参考一面之猿网中的文章内容。相关视频在B站持续更新中，欢迎观看和交流：

【B站视频】CGraph 入门篇
【B站视频】CGraph 功能篇 
- 全面介绍CGraph项目中，所有的名词术语和功能模块
- 结合实际coding过程，详细介绍了每个功能的具体的使用场景、用法、以及解决的问题
- 适合想要全面了解功能和快速上手使用CGraph的童鞋
- 适合对多线程编程感兴趣的童鞋
【B站视频】CGraph 应用篇
【B站视频】CGraph 分享篇

二. 入门Demo

C++ 版本

#include "CGraph.h"

using namespace CGraph;

class MyNode1 : public GNode {
public:
    CStatus run() override {
        printf("[%s], sleep for 1 second ...\n", this->getName().c_str());
        CGRAPH_SLEEP_SECOND(1)
        return CStatus();
    }
};

class MyNode2 : public GNode {
public:
    CStatus run() override {
        printf("[%s], sleep for 2 second ...\n", this->getName().c_str());
        CGRAPH_SLEEP_SECOND(2)
        return CStatus();
    }
};


int main() {
    /* 创建一个流水线，用于设定和执行流图信息 */
    GPipelinePtr pipeline = GPipelineFactory::create();
    GElementPtr a, b, c, d = nullptr;

    /* 注册节点之间的依赖关系 */
    pipeline->registerGElement<MyNode1>(&a, {}, "nodeA");
    pipeline->registerGElement<MyNode2>(&b, {a}, "nodeB");
    pipeline->registerGElement<MyNode1>(&c, {a}, "nodeC");
    pipeline->registerGElement<MyNode2>(&d, {b, c}, "nodeD");

    /* 执行流图框架 */
    pipeline->process();

    /* 清空流水线中所有的资源 */
    GPipelineFactory::remove(pipeline);

    return 0;
}

CGraph Demo 如上图所示，图结构执行的时候，首先执行a节点。a节点执行完毕后，并行执行b和c节点。b和c节点全部执行完毕后，再执行d节点。

Python 版本

import time
from datetime import datetime

from pycgraph import GNode, GPipeline, CStatus


class MyNode1(GNode):
    def run(self):
        print("[{0}] {1}, enter MyNode1 run function. Sleep for 1 second ... ".format(datetime.now(), self.getName()))
        time.sleep(1)
        return CStatus()

class MyNode2(GNode):
    def run(self):
        print("[{0}] {1}, enter MyNode2 run function. Sleep for 2 second ... ".format(datetime.now(), self.getName()))
        time.sleep(2)
        return CStatus()


if __name__ == '__main__':
    pipeline = GPipeline()
    a, b, c, d = MyNode1(), MyNode2(), MyNode1(), MyNode2()

    pipeline.registerGElement(a, set(), "nodeA")
    pipeline.registerGElement(b, {a}, "nodeB")
    pipeline.registerGElement(c, {a}, "nodeC")
    pipeline.registerGElement(d, {b, c}, "nodeD")

    pipeline.process()

其他版本

CsCGraph : A CSharp native, CGraph-API-liked DAG project
JaCGraph : A Java native, CGraph-API-liked DAG project
GoCGraph : A Go native, CGraph-API-liked DAG project
CGraph-lite : A one-header-only, CGraph-API-liked DAG project, lite version by C++

三. 推荐阅读

四. 关联项目

GraphANNS : Graph-based Approximate Nearest Neighbor Search Working off CGraph
CThreadPool : 一个简单好用、功能强大、性能优异、跨平台的C++线程池
PyCGraph-example : A useful list of how cool to use PyCGraph
awesome-cpp : A curated list of awesome C++ (or C) frameworks, libraries, resources, and shiny things. Inspired by awesome-... stuff.
awesome-workflow-engines : A curated list of awesome open source workflow engines
taskflow : A General-purpose Parallel and Heterogeneous Task Programming System
- 【B站视频】CGraph 和 taskflow 性能对比实测
torchpipe : Serving Inside Pytorch
- 【B站视频】开源项目torchpipe - ai引擎在线高并发经验和实战
nndeploy : Easy-to-use, high-performance, multi-platform inference deployment framework
- 【B站视频合集】【nndeploy】AI推理框架最全教程
KuiperInfer : 带你从零实现一个高性能的深度学习推理库，支持大模型 llama2 、Unet、Yolov5、Resnet等模型的推理。Implement a high-performance deep learning inference library step by step
- 【B站视频】KuiperInfer推理框架 - 一个面向教学的推理框架
OGraph : A simple way to build a pipeline with Go.
- [【B站视频】听阿里云大佬分享：OGraph——基于Go的流图调度二三事](https://www.bili

CGraph

Install / Use

README

一. 简介

二. 入门Demo

三. 推荐阅读

四. 关联项目