NVIDIA Merlin is an open source library providing end-to-end GPU-accelerated recommender systems, from feature engineering and preprocessing to training deep learning models and running inference in production.
Apache License 2.0
766
stars
118
forks
source link
[Task] Transformer-session-based End2End Example with Merlin Models and Merlin Systems #694
Description
Use Case Number #2 in https://github.com/NVIDIA-Merlin/models/issues/734 and see https://docs.google.com/document/d/1Zi0SNHHiN72wYIwpkKx-u5xqC35YvlZjNTKUosRuQds/edit