search

openmlsys / openmlsys-zh

《Machine Learning Systems: Design and Implementation》- Chinese Version
https://openmlsys.github.io
3.84k stars 421 forks source link

[拓展章节] Graph + MLSys #231

Open hansen7 opened 2 years ago

hansen7 commented 2 years ago

与@luomai 老师交流后，我很希望能够Co-Lead/参与到这一拓展章节的撰写过程中，我对Graph, GNN等比较熟悉，刚开始了解MLSys + Graph这一有趣的方向。

luomai commented 2 years ago

可以的。我们大概花2个月左右时间写这章。先在这里讨论下章节的内容。

eedalong commented 2 years ago

GNN应用 计算范式: Message Passing 编程范式：Tensor-Centric(PyG), Graph-Centric(DGL), Vertex-Centric(MindSpore-GraphLearning) 整图学习系统： RoC 基于采样子图Mini-Batch学习系统： PyG， DGL

eedalong commented 2 years ago

@luomai @hansen7 本章节需要覆盖的一些内容的proposal～

yzh119 commented 2 years ago

I don't agree that there are differences between the paradigms between DGL, PyG, and Seastar. All of them adopt vertice/edge-centric DSL. The only differences are how is the DSL compiled and executed, and the methodology converges as time went by.

I encourage paying more attention to distributed training of GNNs: how to optimize pipelining, networking, sampling, etc.

luomai commented 2 years ago

@yzh119 谢谢建议。图学习系统这块我们还没有开始梳理出来。很欢迎提建议和参与。我们最近在忙第一版的发布。忙完这阵就会开始思考图学习系统的章节了。

yofufufufu commented 1 year ago

各位老师好，想请问图学习这一块进展如何了？对Graph+MLSys感兴趣，非常期待！