Overview
Explore the differences between Kubeflow and MLFlow in this 43-minute panel discussion featuring AI/ML experts from Canonical Ubuntu. Dive into production-grade and open-source MLOps, comparing these popular machine learning platforms. Learn about their strengths, weaknesses, and use cases to determine which solution best fits your AI and machine learning needs. Gain insights on experiment tracking, model registry, and integrations with other tools like Spark, Grafana, and Prometheus. Understand the challenges in AML, the path to production, model retraining, and the importance of community-driven ML tooling. By the end, you'll have a comprehensive understanding of when to choose Kubeflow or MLFlow for your machine learning operations.
Syllabus
Introduction
AML Challenges
Path to Production
Retraining Models
Open Source ML
Who wins
Kubeflow
MLFlow
MLFlow Overview
Similarities
Controversial part
When to choose
Canonical
Open Source
Integration
Use Cases
Tracking Experiments
Conclusion
Taught by
Canonical Ubuntu