DLRM Workload for Recommendation: PyTorch

mlcommons / algorithmic-efficiency

MLCommons Algorithmic Efficiency is a benchmark and competition measuring neural network training speedups due to algorithmic improvements in both training algorithms and models.

https://mlcommons.org/en/groups/research-algorithms/

Apache License 2.0

329 stars 66 forks source link

DLRM Workload for Recommendation: PyTorch #32

Closed fsschneider closed 2 years ago

fsschneider commented 3 years ago

Workload

Task

Binary Classification

Dataset

Criteo Terabyte dataset

Model

DLRM

Reference Implementation

There is an existing MLCommons Training implementation for PyTorch.

fsschneider commented 3 years ago

For some reason, I couldn't assign or tag Rakshith or Nikola. Just wanted to note down that they are currently working on this.

pomonam commented 2 years ago

The PR https://github.com/mlcommons/algorithmic-efficiency/pull/146 implements this workload.