Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[ASoC 2022] Improve DeepRec ModelZoo. #232

Closed
liutongxuan opened this issue May 25, 2022 · 0 comments
Closed

[ASoC 2022] Improve DeepRec ModelZoo. #232

liutongxuan opened this issue May 25, 2022 · 0 comments

Comments

@liutongxuan
Copy link
Member

Background

This is an basic subject of ASoC 2022 and #231 .

There are 6 models in ModelZoo in DeepRec. Currently, there is only model code for training. Please add inference code for these models and optimize the inference performance, and summarize performance results.

Target

  1. Design and implement inference code for models in ModelZoo.
  2. Profiling and optimize inference performance for models in ModelZoo.

Difficulty

Basic

Mentor

@shanshanpt jiankeng.pt@alibaba-inc.com

Output Requirements

Proficiency in C++ and Python;
Get to know DeepRec;
Able to complete the development under the guidance of the mentor;
Have a certain understanding and interest in deep learning recommendation engines;

背景

这是一个阿里巴巴编程之夏 2022 的基础课题 #231 .

DeepRec中ModelZoo中有6个模型,当前没有支持导出为SavedModel,导致训练和推理不能直接打通。请完善这些模型并且完成训练推理完整链路的测试。

目标

1)实现ModelZoo中6个模型的Inference use case。
2)优化ModelZoo的模型的Inference性能,并总结性能文档。

难度

基础

导师

@shanshanpt jiankeng.pt@alibaba-inc.com

产出要求

熟练掌握C++和Python;
能够在导师的指导下熟悉并理解相关的代码
了解 DeepRec;
对深度学习推荐引擎有一定了解和兴趣;

@liutongxuan liutongxuan changed the title [ASoC 2022] DeepRec ModelZoo Improvement (DeepRec ModelZoo完善) [ASoC 2022] Improve DeepRec ModelZoo. May 25, 2022
marvin-Yu pushed a commit to changqi1/DeepRec that referenced this issue Apr 9, 2023
…15/unique

[TF 1.15] Update Unique kernel with UniqueOpHashMap
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant