Continuous Deployment Platform for ML Models (Cloud Next '18)

Опубликовано 27 июля 2018, 16:05

Applying the machine learning to your business requires a platform for the continuous deployment of ML models. In this session, you will learn key components such as "Training platform", "Validation platform" and "Serving platform", and how to combine them to build an end-to-end platform for the continuous deployment. As a concrete example, we will demonstrate an example architecture consisted of multiple GCP services to explain how data freshness affects your ML models, and why the continuous deployment is critical to keep the quality of your ML models.

MLAI213

Event schedule → g.co/next18

Watch more Machine Learning & AI sessions here → bit.ly/2zGKfcg
Next ‘18 All Sessions playlist → bit.ly/Allsessions

Subscribe to the Google Cloud channel! → bit.ly/NextSub


event: Google Cloud Next 2018; re_ty: Publish; product: Cloud - General; fullname: Etsuji Nakai;