So what exactly are vector databases?

Vector databases are like smart organizers for information. They are important because they help us find things quickly and accurately. Imagine you have a big collection of pictures, and you want to find all the pictures of dogs. Instead of going through each picture one by one, a vector database can analyze the unique features of the images, like the shape of the ears or the color of the fur. This is possible through what we call “Embeddings”, which are a way to measure those features. It then creates special codes called vectors that represent those features. So when you search for “dog,” the database compares your search with all the vectors it has and finds the pictures that are most like dogs. This saves a lot of time and effort! 

But that’s not all! Vector databases can also help us with other things, like finding recommendations. Let’s say you’re listening to music on a streaming platform and you really like a song. The platform can use a vector database to analyze the characteristics of that song, like the rhythm or the melody. Then, it can compare it to other songs in its library and find similar ones that you might enjoy. So, vector databases help us organize and discover things in a clever and efficient way, whether it’s finding pictures or suggesting awesome songs!

In this project you will learn how to create your own vector database using open-source tools. By the end of this guided project, you will know:

However, don’t worry if you don’t have much experience with the above. This project will explain each step along the way, and you may learn as-you-need along the way.

IBM Skills Network Labs environment also comes with many technical requisites pre-installed to save the hassle of setting everything up, so we won’t be dealing with any environment issues.