Discover the new Snowpark

 

Discover the new Snowpark ML Toolkit + dbt Python models

Snowpark ML is a new set of tools for building and deploying machine learning models in Snowflake. The best part is that you get familiar ML constructs (Scikit-Learn, XGBoost, LightGBM, …), paired with all the power, security, and scalability of Snowflake.

Why I’m writing this

In my previous post I shared how dbt and Snowflake pair beautifully to power the new dbt Python models.

The dbt Python models docs offer an in-depth use case of how to do ML prep, cleaning, training, and prediction with dbt Python + Snowpark.

But — those docs don’t leverage the new Snowpark ML Toolkit — which will make these previous tasks perform even better.

Meanwhile the Snowpark ML Toolkit quickstart offers an in-depth explanation of how to use this new ML Toolkit.

But — that quickstart starts by asking us to download and install Python libraries in our VMs. I don’t like that step. I would rather use a cloud tool (like dbt Cloud), and have every library managed “magically”.

Here