Faiss: The Missing Manual
Facebook AI Similarity Search (Faiss) is one of the best open source options for similarity search. In this ebook, you will learn the essentials of vector search and how to apply them in Faiss to build powerful vector indexes.
Introduction
Vector search has been used by tech giants like Google and Amazon for decades. It has been claimed to be a significant driver in clicks, views, and sales across several platforms. Yet, it was only with Faiss that this technology became more accessible.
In the past few years, vector search exploded in popularity. It has driven ecommerce sales, powered music and podcast search, and even recommended your next favorite shows on streaming platforms. Vector search is everywhere and in the following chapters you will discover why it has found such great success and how to apply it yourself using the Facebook AI Similarity Search (Faiss) library.
Introduction to Facebook AI Similarity Search (Faiss)
An overview of the Faiss library and similarity search.
Chapter 2Nearest Neighbor Indexes for Similarity Search
Learn how to choose the right index in Faiss.
Chapter 3Locality Sensitive Hashing (LSH): The Illustrated Guide
Take your first steps to a deeper understanding of approximate nearest neighbor indexes with LSH.
Chapter 4Random Projection for Locality Sensitive Hashing
Apply LSH to modern dense vector representations using random projection.
Chapter 5Product Quantization
Learn how Product Quantization (PQ) can be used to compress indexes by upto 97%.
Chapter 6Hierarchical Navigable Small Worlds (HNSW)
HNSW graphs are among the top performing indexes in similarity search.
Chapter 7Composite Indexes and the Faiss Index Factory
Learn how to apply all we have learned so far to create multi-step composite indexes.
And more...
What will you build?
Upgrade your search or recommendation systems with just a few lines of code, or contact us for help.