Share via

AI and machine learning tutorials

  • Article

The tutorials in this section illustrate how to use Azure Databricks throughout the AI lifecycle for classical ML and gen AI workloads.

If you're new to AI on Azure Databricks, see Try generative AI and machine learning on Databricks for a curated list of notebooks and tutorials designed to quickly get you started with AI.

Classical ML tutorials

You can import each notebook to your Azure Databricks workspace to run them.

Notebook Features
Deploy and query a custom model Unity Catalog, classification model, MLflow, model serving, Hugging Face transformer, PyFunc model
Machine learning with scikit-learn Unity Catalog, classification model, MLflow, automated hyperparameter tuning with Optuna and MLflow
Machine learning with MLlib Logistic regression model, Spark pipeline, automated hyperparameter tuning using MLlib API
Deep learning with TensorFlow Keras Neural network model, inline TensorBoard, automated hyperparameter tuning with Hyperopt and MLflow, autologging, ModelRegistry

Deep learning tutorial

Notebook Requirements Features
End-to-end PyTorch example Databricks Runtime ML Unity Catalog, PyTorch, MLflow, automated hyperparameter tuning with Optuna and MLflow

Gen AI tutorials

You can import each notebook to your Azure Databricks workspace to run them.

Notebook Features
Query OpenAI external model endpoints OpenAI API, MLflow, External models, Databricks Secrets
Create and deploy a Foundation Model Fine-tuning run Foundation Model fine-tuning, databricks_genai SDK
Build, evaluate, and deploy production-grade AI agents Mosaic AI Agent Framework, Agent Evaluation, MLflow, synthetic data