Explore MLFlow, an open source platform for managing machine learning workflows, in this 46-minute webinar presented by Andreea Munteanu from Canonical Ubuntu. Gain insights into MLFlow's position in the machine learning landscape, its benefits, and its four main components: experiment tracking, model registry, model deployment, and code packaging. Discover the differences between the upstream project and Charmed MLFlow, Canonical's distribution. Learn about best use cases for Charmed MLFlow, when to integrate it with other MLOps platforms, and how to build an end-to-end MLOps solution. Delve into topics such as AI readiness, open source in machine learning, MLFlow's history and challenges, MLFlow projects, models, and experiments. Compare MLFlow with Kubernetes and explore its various applications. Conclude with a discussion on Clinical AI Roadshow and a Q&A session.
Introduction to MLFlow: Managing Machine Learning Workflows
Overview
Syllabus
Introduction
AI is everywhere
Challenges
AI Readiness
Open Source
MLFlow
History of MLFlow
Challenges of MLFlow
Why use MLFlow
What is MLFlow
MLFlow Projects
MLFlow Models
Model Registry
Runs and Experiments
How to use MLFlow
MLFlow debate
MLFlow vs Kubernetes
MLFlow uses
Charm MLFlow
Charmed vs Upstream
Questions
Clinical AI Roadshow
Taught by
Canonical Ubuntu
