Custom Training is a do-it-yourself solution to build an ML project.

Before any coding begins, you must determine what environment you want your ML training code to use.

• pre-built container: container that comes with many off-the-shelf solutions for general machine learning practice
• custom container: comes with no pre-defined tools; you must determine the details like the environment, machine type, and disks when creating the custom container

In terms of the tools to code your ML model, you can use Vertex AI Workbench. It is like a Jupyter notebook deployed in a single environment that supports the entire data science workflow, from exploring to training and then deploying a machine learning model.

You can also use Colab Enterprise.

Vertex AI fully hosts TensorFlow from low-level to high-level APIs.