Accelerating the Wide & Deep Model Workflow from 25 Hours to 10 Minutes Using NVIDIA GPUs

Originally published at: Accelerating the Wide & Deep Model Workflow from 25 Hours to 10 Minutes Using NVIDIA GPUs | NVIDIA Developer Blog

Recommender systems drive engagement on many of the most popular online platforms. As data volume grows exponentially, data scientists increasingly turn from traditional machine learning methods to highly expressive, deep learning models to improve recommendation quality. Often, the recommendations are framed as modeling the completion of a user-item matrix, in which the user-item entry is…

The fast preprocessing and model training all done on GPU now enables data scientists to include feature engineering and selection into a normal HP search for the model. This is crucial for recommenders. If you have any questions or comments, let us know.