Skip to main content

AI and machine learning tutorials

Try one of these tutorials to get started. You can import these notebooks to your Databricks workspace.

Tutorial

Description

Classic ML

End-to-end example of training a classic ML model in Databricks.

scikit-learn

Use one of the most popular Python libraries for machine learning to train machine learning models.

MLlib

Examples of how to use the Apache Spark machine learning library.

Deep learning using PyTorch

End-to-end example of training a deep learning model in Databricks using PyTorch.

TensorFlow

TensorFlow is an open-source framework that supports deep-learning and numerical computations on CPUs, GPUs, and clusters of GPUs.

Mosaic AI Model Serving

Deploy and query a classic ML model using Mosaic AI Model Serving.

Foundation model APIs

Foundation model APIs provide access to popular foundation models from endpoints that are available directly from the Databricks workspace.

Agent framework quickstart

Use Mosaic AI Agent Framework to build an agent, add a tool to the agent, and deploy the agent to a Databricks model serving endpoint.

Trace a GenAI app

Trace an app's execution flow with visibility into every step.

Evaluate a GenAI app

Use MLflow 3 to create, trace, and evaluate a GenAI app.

Human feedback quickstart

Collect end-user feedback and use that feedback to evaluate your GenAI app's quality.

Build, evaluate, and deploy a retrieval agent

Build an AI agent that combines retrieval with tools.

Query OpenAI models

Create an external model endpoint to query OpenAI models.