GoogleCloudPlatform / vertex-pipelines-end-to-end-samples

Apache License 2.0

232 stars 93 forks source link

Extending model monitoring config #73

Closed lily-nicholls closed 10 months ago

lily-nicholls commented 10 months ago

Description

Please provide a description of the changes you have made in this PR.

Adding notification channels and enable logging to the model monitoring config in the batch prediction job

How has this been tested?

Please explain how you have tested the new changes.

I have ran prediction pipelines and creating a notification channel for alerts to be sent to

Checklist

[ ] I have commented my code, particularly in hard-to-understand areas
[ ] I have successfully run the E2E tests, and have included the links to the pipeline runs below
[ ] I have added tests that prove my fix is effective or that my feature works
[ ] New and existing unit tests pass locally with my changes
[ ] I have updated any relevant documentation to reflect my changes
[ ] I have assigned a reviewer and messaged them

Pipeline run links:

https://console.cloud.google.com/vertex-ai/pipelines/locations/europe-west2/runs/turbo-prediction-pipeline-20240110163251?project=dt-lily-sandbox-2-dev