Creating a Deployment

Deployments can be created in a few simple steps. Head to Deployments to create a Deployment from the UI. The default deployment flow uses KServe as a deployment backend. If you want to use AWS Sagemaker as a deployment backend you can learn more in this in this article.

As an alternative, deployments can be created using the Deeploy Python Client. We highly recommend this for large Repositories to speed up the process. 

Prerequisites

• You added a Repository that adheres to the requirements

Step 1: Add a repository, and select a version (branch, commit)

Choose to use an existing Repository. Deeploy will list currently linked Repositories in a dropdown. Select the repository you want to use for your Deployment. 

After selecting a repository, Deeploy will list the branches and commits. Select the branch and commit that you want to deploy.

Step 2: Add deployment metadata

In this step, you can add deployment metadata. Only the name is required, the other metadata is optional. All metadata can be added or adjusted after deploying. 

Step 3: Select the model framework or Custom Docker Image

Select the model framework that you have used to train the model as shown in the figure below.  Supported frameworks, versions, and examples can be found in Supported Framework Versions for KServe.

Alternatively, deploy a Custom Docker Images, this article guides you through the steps of deploying your Custom Docker Image. The only additional requirement is to use a webserver to expose an endpoint. 

Indicate whether or not your model should be deployed Serverless. Consult our guide for Serverless deployments if you are unsure if serverless deployments are necessary. 

For more information on the advanced and serverless configuration see:

• Advanced configuration
• Serverless configuration

Step 4: Select the explainer 

Select the explainer framework that you have used to train the explainer as shown in the figure below.  Supported frameworks, versions, and examples can be found in Supported Framework Versions for KServe.

To decide whether or not your explainer should be deployed Serverless, please consult our guide for Serverless deployments.

For more information on the advanced and serverless configuration see:

• Advanced configuration
• Serverless configuration

Initiate the automated deployment 

Click Deploy, Deeploy will now initiate the automated deployment process. Deeploy will show when the deployment is finished.

Up next:  Test a Deployment