Manage model monitor

Preview

This product or feature is subject to the "Pre-GA Offerings Terms" in the General Service Terms section of the Service Specific Terms. Pre-GA products and features are available "as is" and might have limited support. For more information, see the launch stage descriptions.

This page describes how to update and delete model monitors.

Update a model monitor

Update a model monitor to modify the default configuration for monitoring jobs. You can't change the model version that a model monitor is associated with. Instead, create a new model monitor.

Console

1. In the Google Cloud console, go to the Monitoring page.

   Go to Monitoring

2. Click the model monitor to go to the Monitor details page.

3. Click Edit configuration to update the model monitor.

4. Modify your schema, notifications, training data source, and objectives.

5. Click Save.

Python SDK

You can update the following specifications:

• Display name
• Training dataset
• Model schema
• Default monitoring objectives
• Default monitoring output specification
• Default monitoring notification specification
• Default explanation specification for feature attribution

The following example updates the model schema:

monitor_2.update(model_monitoring_schema=MODEL_MONITORING_SCHEMA)

Delete a model monitor

Remove model monitors to prevent authorized users from inadvertently running monitoring jobs on unused model monitors. When you delete a model monitor, any associated monitoring data is deleted except for metrics and statistics exported by Monitoring, which remain in Cloud Monitoring and in your Cloud Storage.

Console

1. In the Google Cloud console, go to the Monitoring page.

   Go to Monitoring

2. Click the model monitor to go to the Monitor details page.

3. Click Edit configuration to update the model monitor.

Python SDK

from google.cloud.aiplatform.private_preview.centralized_model_monitoring
import model_monitor

my_model_monitor.delete(force=true)