In database Machine Learning with Python in SQL Server
summary
Claims that Python runs where the data lives, using SQL Server 2017 machine learning services to train and serve models without data movement. Uses RevoScalePy and Microsoft ML package, trains random forest and boosted trees against full hospital records, finds random forest wins, writes model binaries into a SQL table and scores via Python-based models embedded as stored procedures called from Power BI. Execution runs in isolated external runtimes with low‑privilege accounts and no extra licensing cost; demo completes and dashboards update.
AI-written summary · about
chapters (11)
- 0:00Introduction to in-database machine learning
- 1:24Why in-database ML matters
- 8:44Python in SQL Server setup
- 13:56Scalable ML libraries
- 18:15End-to-end workflow
- 20:02Demo: hospital stay prediction
- 28:35Licensing & deployment
- 30:22External data handling
- 33:13BI tools for predictions
- 34:17Security for database administrators
- 36:52Next steps & resources
related talks
37mAccelerating Python Analytics by In-Database Processing
21mMachine Learning in your Data Warehouse using Python
25mSql Server and R for Real-time Predictions
33mYour best Bet: Effortless MLOps with Python Models in
27mMachine Learning in the Warehouse with Python
40mIn-Database Machine Learning with Jupyter
26mPython + MPP Database = Large Scale AI/ML Projects in Production Faster
29mScaling Python-Powered Machine Learning With Snowflake