ML Ops Best Practices on Google Cloud (Cloud Next '19)

Опубликовано 11 апреля 2019, 1:36

Creating an ML model is just a starting point. To bring it into production, you need to solve various real-world issues, such as building a pipeline for continuous training, automated validation of the model,  scalable serving infrastructure, and supporting multiple environments in increasingly common hybrid and multi-cloud setups. In this session, we will learn the concept of ""ML Ops"" (DevOps for ML) and how to leverage various Google initiatives like TFX, Kubeflow Fairing (Hybrid ML SDK) and Kubeflow Pipelines to build and maintain production quality ML systems.

ML Ops and Kubeflow Pipelines → bit.ly/2KK5hwi

Watch more:
Next '19 ML & AI Sessions here → bit.ly/Next19MLandAI
Next ‘19 All Sessions playlist → bit.ly/Next19AllSessions

Subscribe to the G Suite Channel → bit.ly/G-Suite1

Speaker(s): Kaz Sato, Zia Syed, Robin Zondag, Fabien Da Silva

Session ID: MLAI101
product:Cloud ML Engine,Cloud AI,AI,TensorFlow; fullname:Kaz Sato,Zia Syed; event: Google Cloud Next 2019; re_ty: Publish; product: Cloud - General; fullname: Kaz Sato;