You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
At present, DeepRec cannot support the evaluation of very large models (single node cannot be loaded), multiple PS are required to load large models, and multiple workers are used for distributed evaluation.
Target
Design and implement the capability large model evaluation, support multiple PS loading large model.
Design and implement multiple evaluator node in one job.
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;
Background
This is an advance subject of ASoC 2022 and #231 .
At present, DeepRec cannot support the evaluation of very large models (single node cannot be loaded), multiple PS are required to load large models, and multiple workers are used for distributed evaluation.
Target
Difficulty
Advance
Mentor
@candyzone candy.dc@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 支持多evaluator评估:目前DeepRec下无法支持超大模型(单节点无法加载)的评估,需要多个ps加载大模型,并且使用多worker进行分布式评估。
目标
1)支持超大模型通过多PS方式加载模型,实现Evaluation.
2)支持一个任务中使用多个Evaluator节点进行评估。
难度
进阶
导师
@candyzone candy.dc@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;
The text was updated successfully, but these errors were encountered: