Explore how Kubernetes can streamline machine learning workflows in this 39-minute conference talk by Brian Redmond from Microsoft. Learn about implementing ML solutions on Kubernetes with containers, covering stages like data preparation, model training, testing, validation, monitoring, and CI/CD automation. Discover tools such as Tensorflow/Kubeflow, Pachyderm, and Argo through practical demonstrations. Gain insights into improving efficiency for both data scientists and infrastructure teams by applying DevOps principles to AI and machine learning projects. Delve into topics including MLOps, pipelines, Docker containers, Azure storage, virtual nodes, Helm installation, and Kubernetes instances.
Machine Learning Made Easy on Kubernetes - DevOps for Data Scientists
Overview
Syllabus
Introduction
Demo
Machine Learning
Kubernetes
MLOps
Pipelines
Demonstration
Code
Docker Container
Tensorflow
Deploying TensorFlow
Do I want my data scientists
Cubeflow dashboard
Azure storage
Service
Service IP
Curl
Web App
Game of Thrones
Hyperparameters
Virtual Nodes
Container Instances
Install Helm
Tensor Board
Kubernetes Instances
Khateeb
Kubernetes Pipelines
Training Logs
Notebooks
Demo QA
Taught by
Linux Foundation
