Abstract: We are building a unified AML Kubernetes native agent that will allow you to seamless train ML models on Kubernetes. Supported frameworks include SciKit, TensorFlow , PyTorch and MPI.
The Kubernetes cluster can be AKS in the Azure cloud or it can be on-prem Kubernetes (including Azure Stack Hub, Edge) that you bring to Azure through Azure Arc for Kubernetes. This allows you to centrally manage and govern all your Kubernetes resources in one place and use capacity flexibly for all workloads including AML.
Bio: Saurya Das is a PM in the Azure ML organization focused on the Hybrid strategy and Kubernetes.