Skip to main content
Version: v1.0.0

Workflow

Overview

The typical H2O MLOps workflow can be summarized in the following sequential steps:

In the below sections, each step above, in turn, is summarized.

Step 1: Select the workspace

To begin, select your workspace from the workspaces drop-down in the top navigation bar.

Select workspace

Step 2: Add a model

After selecting a workspace, go to the Models tab in H2O MLOps and add your machine learning model.

The first model you add becomes the initial version. You can add more versions later. Add model versions

Step 3: Deploy the model version

After adding a model, create a deployment for a model version so you can score and monitor its performance. Deploy model version

Step 4: Score against the deployment

After deploying the model, you can run quick scoring on the deployment.

  • To learn more about quick scoring, see Quick scoring.
  • To learn how to score against deployments using H2O MLOps Python client, see Deployment scorer.

Step 5: Monitor the deployed model

After scoring, monitor the deployed model to track its performance and detect issues such as model drift.

note

You must enable and configure model monitoring when you create the deployment.

  • To learn more about model monitoring, see Model monitoring.
  • To learn how to configure monitoring for your deployment using the H2O MLOps Python client, see Monitoring setup.

Feedback