search

DeepRec-AI / HybridBackend

A high-performance framework for training wide-and-deep recommender systems on heterogeneous cluster
Apache License 2.0
152 stars 30 forks source link

DLRM model on A100 8cards training #95

Closed zhaozheng09 closed 1 year ago

zhaozheng09 commented 1 year ago

I don't find a DLRM example in HybridBackend repo. Can you provid a DLRM example ?

2sin18 commented 1 year ago

Hi, there are already DLRM examples for training: https://github.com/alibaba/HybridBackend/blob/main/docs/tutorial/ranking/criteo/train.py

Attention: input dataset is not divided into different slices on different workers in these examples, which would be modified in near future.