Vertex AI Model Garden

(1 day)

 

Vertex AI Model Garden provides enterprise-ready foundation models, task-specific models, and APIs. Model Garden can serve as the starting point for model discovery for various different use cases. You can kick off a variety of workflows including using models directly, tuning models in Vertex AI Studio, or deploying models to a data science notebook.

In this course, after being introduced to Vertex AI as a machine learning platform through the lens of Model Garden, you will learn how to leverage pre-trained models as part of your machine learning workflow and how to fine-tune models for your specific applications.

Course Objectives

  • Understand the model options available within Vertex AI Model Garden
  • Incorporate models in Vertex AI Model Garden in your machine learning workflows
  • Leverage foundation models for generative AI use cases
  • Fine-tune models to meet your specific needs

Audience

Machine learning practitioners who wish to leverage models available in Vertex AI Model Garden for various different use cases.

Prerequisites

To get the most out of this course, participants should have:

  • Prior completion of the Machine Learning on Google Cloud course or the equivalent knowledge of TensorFlow/Keras and machine learning.
  • Experience scripting in Python and working in Jupyter Notebook to create machine
    learning models.

Course Outline

Module 1: Vertex AI for ML Workloads

  • Vertex AI on Google Cloud
  • Options for training, tuning, and deploying ML models on Vertex AI
  • Generative AI options on Google Cloud and Vertex AI

Module 2: Model Garden

  • Introduction to Model Garden
  • Model types in Model Garden
  • Connecting models from Vertex AI Studio and Model Registry
  • Introduction to course use cases

Module 3: Task-Specific Solutions: Content Classification

  • Pre-trained models for specific tasks
  • VertexAI AutoML
  • Using a pre-trained model via the Python SDK
  • Lab: Content Classification via Natural Language API and AutoML

Module 4: Foundation Models: Text Embeddings via PaLM

  • Introduction to foundation models
  • PaLM API
  • Vertex AI Studio
  • Using the Embeddings API
  • Lab: Use the PaLM API to Cluster Products Based on Descriptions

Module 5: Fine-Tunable Models

  • Fine-tunable models in Model Garden
  • Vertex AI Pipelines
  • Demo: Fine-Tuning Models for Your Specific Use Case