Deployment

A Deployment is an instance of a hosted machine learning model. It refers to a physical deployment on the underlying hardware.

• Next to hosting the model, a Deployment can (optionally) also host an explainer.
• Every Deployment has an endpoint that can be used to communicate with the model to fetch predictions and explanations.
• Every Deployment has a Deployment owner. The owner represents the user account that is responsible for the Deployment. The initial Deployment owner is the user that creates the deployment.

More detailed information about creating deployments can be found here.

The concept of a deployment is closely related to the concept of a Repository.