NGC | Catalog
Welcome Guest
CatalogModelsDLRM checkpoint (PyTorch, AMP, BS64k, Base, FL15)

DLRM checkpoint (PyTorch, AMP, BS64k, Base, FL15)

For downloads and more information, please view on a desktop device.
Logo for DLRM checkpoint (PyTorch, AMP, BS64k, Base, FL15)

Description

DLRM PyTorch checkpoint trained on Criteo Dataset with FreqLimit=15 on A100 with AMP

Publisher

NVIDIA Deep Learning Examples

Use Case

Recommendation

Framework

PyTorch

Latest Version

21.02.0

Modified

October 29, 2021

Size

15.61 GB

Model Overview

The Deep Learning Recommendation Model (DLRM) is a recommendation model designed to make use of both categorical and numerical inputs.

Model Architecture

DLRM accepts two types of features: categorical and numerical. For each categorical feature, an embedding table is used to provide dense representation to each unique value. The dense features enter the model and are transformed by a simple neural network referred to as "bottom MLP". This part of the network consists of a series of linear layers with ReLU activations. The output of the bottom MLP and the embedding vectors are then fed into the "dot interaction" operation. The output of "dot interaction" is then concatenated with the features resulting from the bottom MLP and fed into the "top MLP" which is also a series of dense layers with activations. The model outputs a single number which can be interpreted as a likelihood of a certain user clicking an ad.


Figure 1. The architecture of DLRM.

Training

This model was trained using script available on NGC and in GitHub repo

Dataset

The following datasets were used to train this model:

  • Criteo 1TB - Dataset containing feature values and click feedback for millions of display ads. Its purpose is to benchmark algorithms for clickthrough rate (CTR) prediction.

Performance

Performance numbers for this model are available in NGC

References

License

This model was trained using open-source software available in Deep Learning Examples repository. For terms of use, please refer to the license of the script and the datasets the model was derived from.