Enabling Machine Learning Operations
Enable Machine Learning Operations (MLOps) so that you can replace active models in predictions with other models, scorecards, or fields, and then deploy the candidate models to production. To enable MLOps, update Prediction Studio settings with the work queue for data scientists, the analytics repository, and optionally, an email account for sending notifications.
- In the navigation pane of Prediction Studio, click .
- Optional: If you want to use a different work queue for data scientists than the default
work queue DataScientistWorkqueue, in the Work
queue field, enter the name of your work queue.For more information, see Creating a work queue.
- Add your data scientists' operator IDs to the work group that is associated
with the work queue configured in the Work queue
Note: For example, the default work queue DataScientistWorkqueue is configured with the Default work group. To received work related to model updates, the operator must be a part of the Default work group.
For more information, see Defining work routing settings for an operator.
- In the Storage section, in the Analytics
repository field, enter the name of the analytics repository for
storing machine learning models.
For more information, see Specifying a repository for Prediction Studio models.
- Click Save.
Previous topic Analyzing example projects and models in Prediction Studio Next topic Configuring the monitoring of model input and output