Implementing MLOps on Google Cloud using Vertex AI Pipelines
In this section, we will cover the steps to build an MLOps pipeline on Google Cloud using Vertex AI Pipelines.
Prerequisite: IAM permissions
In this section, we will use the same Vertex AI Workbench notebook instance that we created in Chapter 5. That user-managed notebook uses the default Compute Engine service account, which is granted the IAM basic Editor role by default. When we build and execute our pipeline in our notebook, we decide to let our pipeline inherit the same permissions used by our notebook. This is a decision we make by not specifying a different role for our pipeline executions. We could specify a different role if we wanted to, but for simplicity purposes, we’ll take the approach of having our pipeline use the default Compute Engine service account. By default, the Editor role (used by the default Compute Engine service account) will allow us to perform all of the required activities in Vertex...
