Machine Learning in the Warehouse with Python
summary
Demonstrates that Snowpark for Python lets notebooks prototype, train, save and run models without exporting data: code registers a local function and a conda-style environment, the model binary lands in a Snowflake stage, predictions run over table rows via SQL-backed Python user-defined functions, and results get written back to a new table. Demo uses pandas and scikit-learn inside Deepnote, and notes limits on training millions of rows without provisioning larger Snowflake resources.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:11Snowpark and Deepnote integration
- 3:59SQL without writing SQL
- 7:57Demo setup
- 10:28Exploratory data analysis
- 15:03Prototyping locally
- 16:50Deploying models
- 18:16UDFs for predictions
- 21:30Snowpark ML workflow
- 23:16Deepnote EDA resources
- 24:17Q&A
related talks
21mMachine Learning in your Data Warehouse using Python
29mScaling Python-Powered Machine Learning With Snowflake
32mScalable Python and SQL Data Engineering without Migraines
23mAcceleratinag Data Science and Machine Learning with Snowflake DataTech
47mData Science and Machine Learning with Python
37mIn database Machine Learning with Python in SQL Server
04h 02mHands-on Session on Data Science for Beginners
21mUsing DBT and Python to Build Production Pipelines in Snowflake