What you'll learn

Building machine learning models on Databricks requires more than training algorithms in a notebook; teams also need scalable data preparation, experiment tracking, model governance, tuning, evaluation, and prediction workflows.

In this course, Building Machine Learning Models on Databricks, you’ll gain the ability to develop, track, tune, register, and evaluate machine learning models using the Databricks machine learning environment.

First, you’ll explore Databricks Runtime for Machine Learning, ML compute, and MLflow Tracking.

Next, you’ll discover how to prepare data with Spark DataFrames, build feature pipelines, train XGBoost models, register models with Unity Catalog governance, and deploy a model for prediction using an API endpoint.

Finally, you’ll learn how to perform hyperparameter tuning using Optuna, compare tuning runs with MLflow, run scalable predictions, evaluate model performance, and train deep learning models with TensorFlow on Databricks.

When you’re finished with this course, you’ll have the skills and knowledge of machine learning on Databricks needed to build governed, scalable, and trackable ML workflows from data preparation through model evaluation.