Kubeflow on AWS (in company)

Overview

Kubeflow is a framework for running Machine Learning workloads on Kubernetes. TensorFlow is a machine learning library and Kubernetes is an orchestration platform for managing containerized applications.

This instructor-led, live training (online or onsite) is aimed at engineers who wish to deploy Machine Learning workloads to an AWS EC2 server.

By the end of this training, participants will be able to:

• Install and configure Kubernetes, Kubeflow and other needed software on AWS.
• Use EKS (Elastic Kubernetes Service) to simplify the work of initializing a Kubernetes cluster on AWS.
• Create and deploy a Kubernetes pipeline for automating and managing ML models in production.
• Train and deploy TensorFlow ML models across multiple GPUs and machines running in parallel.
• Leverage other AWS managed services to extend an ML application.

Format of the Course

• Interactive lecture and discussion.
• Lots of exercises and practice.
• Hands-on implementation in a live-lab environment.

Course Customization Options

• To request a customized training for this course, please contact us to arrange.

Requirements

• An understanding of machine learning concepts.
• Knowledge of cloud computing concepts.
• A general understanding of containers (Docker) and orchestration (Kubernetes).
• Some Python programming experience is helpful.
• Experience working with a command line.

Audience

• Data science engineers.
• DevOps engineers interesting in machine learning model deployment.
• Infrastructure engineers interesting in machine learning model deployment.
• Software engineers wishing to integrate and deploy machine learning features with their application.

Course Outline

• Introduction
• Kubeflow on AWS vs on-premise vs on other public cloud providers
• Overview of Kubeflow Features and Architecture
• Activating an AWS Account
• Preparing and Launching GPU-enabled AWS Instances
• Setting up User Roles and Permissions
• Preparing the Build Environment
• Selecting a TensorFlow Model and Dataset
• Packaging Code and Frameworks into a Docker Image
• Setting up a Kubernetes Cluster Using EKS
• Staging the Training and Validation Data
• Configuring Kubeflow Pipelines
• Launching a Training Job using Kubeflow in EKS
• Visualizing the Training Job in Runtime
• Cleaning up After the Job Completes
• Troubleshooting
• Summary and Conclusion

.

Onze on line trainingen worden door een live instructeur verzorgd.

• Onze DaDesktop® -technologie creeert een digitale leeromgeving (en indien nodig een geclusterde enterprise infrastructuur) waarmee opdrachten en oefeningen uitgevoerd kunnen worden.
• De deelnemers (en de trainer) hebben toegang tot deze virtuele leeromgeving via de browser zodat hij/zij oefeningen kan doen die real time ingezien kunnen worden door de trainer.
• De trainer monitort niet alleen de voortgang van de prakitische oefeningen maar kan ook te hulp schieten en ingrijpen mocht dat nodig zijn.
• Onze remote trainingen verschillen niet van onze klassikale cursussen zowel qua duur, interactiviteit, praktische oefeningen als het cursusmateriaal.
• Door de flexibiliteit in de trainingsopbouw, de hoge mate van interactie tussen trainer en deelnemer en de hands-on oefeningen zijn onze onlinetrainingen zeer efficiënte en effectief.
• Ook onze in-company trainingen leveren wij on line met live instructeur.