How to build ML Models in Snowflake using Python UDFs and Snowpark

Supporting code for the Snowflake BUILD 2022 session "How to build ML Models in Snowflake using Python UDFs and Snowpark".

Contributors

• Sonny Rivera
• Chris Hastie

Snowflake Python Functionality and Setup

To simplify creating Snowpark sessions for the enclosed scripts, a custom module called "interworks_snowpark" has been used. More details can be found in the InterWorks Snowflake Python Functionality GitHub repository. This repository also contains instructions on how to configure your local environment for Snowpark for Python, and pairs well with this Definitive Guide to Snowflake Sessions with Snowpark for Python.

Jupyter Notebooks through VSCode

If you are using VSCode to view the Jupyter notebooks using the Jupyter extension, be sure to also install the ipykernel module into your Python virtual environment

pip install ipykernel

Acknowledgments

Acknowledgements here