In-Database Machine Learning with Jupyter
summary
Shows a Jupyter + Python workflow that keeps data and compute inside an MPP analytical database using verticaPy to run joins, interpolation and training without pulling data to local memory. Example builds a smart‑meter pipeline joining meter, weather and device tables, finds 5 clusters with k‑means, trains a regression and saves the model as PML, and claims deployment takes minutes not months on production‑scale data with hundreds of concurrent users.
AI-written summary · about
chapters (12)
- 0:00Introduction to talk
- 1:03Why in-database ML matters
- 4:16Capabilities of modern MPP databases
- 7:29Problems with traditional ML workflows
- 10:20Advantages of distributed ML in databases
- 14:15Concurrency and scalability benefits
- 16:02Database ML feature summary
- 18:49Python and Jupyter integration
- 19:30Hands-on Jupyter notebook demo
- 37:57Model deployment and production
- 39:02Real-world success story
- 40:05Q&A and resources
related talks
26mPython + MPP Database = Large Scale AI/ML Projects in Production Faster
55mDistributed In-Database Machine Leaning
46mShorten the Path to Production with In-Database Machine Learning
37mAccelerating Python Analytics by In-Database Processing
31mIn-database machine learning: Automate, scale, deploy
28mCharge Your Database with Predictive ML Capabilities using Open-Source
37mIn database Machine Learning with Python in SQL Server
55mGetting Started with Spark and ML