Vertex AI Pipelines
Vertex AI Pipelines allow you to automatically orchestrate your ML workflow in a serverless manner using TensorFlow Extended (TFX) or Kubeflow. Each Vertex AI pipeline job is generated from a configuration file that outlines a list of steps. A typical Vertex AI pipeline imports data into a dataset, trains a model using a training pipeline, and deploys the model to a new endpoint for prediction. Pipeline jobs are run using compute resources, with the following options:
- You can write custom configurations for pipeline jobs using the Kubeflow DSL.
- You can create, run, and schedule pipeline jobs.
- You can specify Service Account or use Compute Default Service Account if not specified.
Google Vertex AI Pipelines orchestrates your ML workflow, based on your descriptions of the workflow as a pipeline. ML pipelines are portable and scalable ML workflows that are based on containers. ML pipelines are composed of a set of input parameters and a list...
