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.
Pull requests from external contributors require approval from a NVIDIA-Merlin organization member with write permissions or greater before CI can begin.
Added 'RAPIDS' prefix for the cuDF mention to refer the overarching project.