ML workflow orchestration
As we discussed in Chapter 7, Exploring Google Cloud Vertex AI, Vertex AI Pipelines is a fully managed service that allows you to retrain your models as often as necessary so that you can adapt to changes and maintain performance over time. We recommend Vertex AI Pipelines for Cloud ML workflow orchestration.
If you’re using the Google TensorFlow framework, we recommend using TensorFlow Extended to define your pipeline and the operations for each step, then executing it on Vertex AI’s serverless pipeline system. TensorFlow provides pre-built components for common steps in the Vertex AI workflow, such as data ingestion, data validation, and training.
If you are using other frameworks, we recommend using Kubeflow Pipeline, which is very flexible and allows you to use simple code to construct pipelines. Kubeflow Pipeline also provides Google Cloud pipeline components such as Vertex AI AutoML.
