In this module, you focused on the two tools in Vertex AI to develop NLP projects: AutoML and custom training.

You started by reviewing the three options provided by Google Cloud to develop an NLP project:

• Pre-built APIs, such as the Dialogflow API, use pre-built NLP models and do not require any training data.

• AutoML is a no-code solution to build a custom NLP model with only click-and-point.

• Custom training with Vertex AI Workbench is a code-based solution where you have full control over the environment and the training process.

You were then introduced to Vertex AI, a unified platform that brings all the components of the machine learning ecosystem and workflow together, to solve production and ease-of-use challenges.

After that, you explored AutoML which is short for automated machine learning, and understood the three major NLP tasks that can be performed with AutoML, including

• text classification,

• entity extraction, and

• sentiment analysis.

You also explored custom training with Vertex AI Workbench, which is a notebook tool and a single development environment for the entire NLP workflow.

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

You then walked through an end-to-end workflow to build an NLP project from data preparation, to model training, and to model serving.

Finally, you put all the knowledge together and practiced in a hands-on lab to decide the article source by using AutoML text classification.