Data + AI Summit 2022: June 27–30, Moscone Center + Virtual  Image

Data + AI Summit 2022: June 27–30, Moscone Center + Virtual

Join us for unique community-driven content, deep technical training and outstanding speakers like Andrew Ng and Ali Ghodsi.

Machine Learning

A data-native and collaborative ML solution for the full ML lifecycle

Try for Free Schedule a demo

background-image

Built on an open lakehouse architecture, Databricks Machine Learning empowers ML teams to prepare and process data, streamlines cross-team collaboration, and standardizes the full ML lifecycle from experimentation to production.

Open Data Lakehouse Diagram

MLflow prediction metrics Diagram

Simplify all aspects of data for ML

Because Databricks ML is built on an open lakehouse foundation with Delta Lake, you can empower your machine learning teams to access, explore and prepare any type of data at any scale. Turn features into production pipelines in a self-service manner without depending on data engineering support.

Automate experiment tracking and governance

Managed MLflow automatically tracks your experiments and logs parameters, metrics, versioning of data and code, as well as model artifacts with each training run. You can quickly see previous runs, compare results and reproduce a past result, as needed. Once you have identified the best version of a model for production, register it to the Model Registry to simplify handoffs along the deployment lifecycle.

Automate  experiment example

Activities example

Manage the full model lifecycle with the Model Registry

Once trained models are registered, you can collaboratively manage them through their lifecycle with the Model Registry. Models can be versioned and moved through various stages, like experimentation, staging, production and archived. The lifecycle management integrates with approval and governance workflows according to role-based access controls. Comments and email notifications provide a rich collaborative environment for data teams.

Deploy ML models at scale and low latency

From the Model Registry, quickly deploy production models using batch scoring for scale, or Databricks Serving for low-latency online serving as REST API endpoints. Because the Model Registry relies on the MLflow Model format, it benefits from ecosystem integrations for a wide variety of deployments, like deploying Docker containers on Kubernetes or loading a model onto a device.

MLflow Model registry Diagram

background-image

Product components

AstraZeneca

Customer Story

AI is transforming drug discovery

AstraZeneca leverages data and NLP to help scientists research novel drugs

HSBC

Customer Story

Reinventing mobile banking with ML

How HSBC manages cashless transactions for over 39M customers with ML-based fraud detection

Starbucks

Customer Story

Brewing data and AI at scale

Starbucks serves up omnichannel experiences across 30,000 stores with Databricks

Comcast

Customer Story

The future of entertainment with AI

Comcast delivers an Emmy-winning viewer experience with AI-powered voice remotes

Ready to get started?