In this section, you’ll advance to custom training and explore the options that Vertex AI provides to build the development environment.

If you don’t want to hand over everything to AutoML but rather wish to have full control of the NLP training and deployment, you can choose custom training with Vertex AI Workbench.

Vertex AI Workbench is a notebook tool, with which you can code with your favorite libraries such as TensorFlow and your favorite languages such as Python.

Vertex AI Workbench is also a single development environment to support an end-to-end NLP workflow,

starting from uploading data,

to model training,

and to model deployment.

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

There are two options: a pre-built container or a custom container.

Imagine that a container is a room.

A pre-built container would represent a fully furnished room with cabinets and appliances.

Cabinets could represent dependencies in the NLP development environment,

and appliances could represent the libraries that you need to build an NLP model.

So, if your NLP training needs a platform like

TensorFlow, PyTorch, Scikit-learn, or XGBoost and a Python code to work with the platform,

a pre-built container is probably your best solution.

A custom container, alternatively, is like an empty room with no cabinets nor appliances.

You define the exact tools that you need to complete the job.