Skip to main content

101M Work with: Multimodal data

Course overview

In this project-based course, you will learn how to work with multimodal data using Weaviate and a movie dataset.

You will get hands-on experience on how to store and index text and image data to be searchable together by meaning, using Weaviate's vectorization capabilities. You will learn how to search through that data using multimodal search methods, as well as filters. You will also learn how to use Weaviate's retrieval augmented generation (RAG) capabilities to generate outputs based on the retrieved objects.

Learning objectives

  Here, we will cover:

Learning Goals
  • How to create a local Weaviate instance, add data to it to enable multi-modal searching, and use AI through retrieval augmented generation.

  By the time you are finished, you will be able to:

Learning Outcomes
  • Create a local instance of Weaviate with a multimodal vectorizer module
  • Produce, store and index multimodal data
  • Perform multimodal searches
  • Use AI (large language models) to augment and transform retrieved data

Units

1. Weaviate for multimodal data

Practical

Create a local Weaviate instance for multimodal data.

2. Populate the database

Practical

Create a collection and import multimodal data, and have Weaviate create vectors for you.

3. Perform searches

Practical

Learn how to use search functions in Weaviate.

4. LLMs and Weaviate (RAG)

Practical

Use large language models to augment and transform retrieved data.