Calling a Transformer ML Model directly via SQL to predict sentiments

Tutorial on applying a Hugging Face Machine Learning model directly to some table data via SparkSql UDFs and MLflow

MLflow and Apache Spark shine for manipulating your data. Let’s focus on showing how an already existing ML model, here the popular distilbert, can be made available in SQL.

Democratize your ML: SQL is much simpler to use than regular Python, what if the model was easily available to your SQL user base?

Applying prediction directly on columns in a table

Here is the high level architecture:

Registering the model into MLflow then Spark

First we pull the model from the Hugging Face and register it in MLflow:

Then via this Notebook we demo how to make the model available as a function that can directly be called in SQL queries.

The code is available in this demo repository. Next time we will see how to build our own model!

And that’s it, happy predicting!