May 22
Loading video player...
In this video, we break down vector search: the idea behind search systems that find meaning, not just matching keywords. We start with a simple problem: what happens when a user asks one thing, but the best document says it in completely different words? I’ll show how vector search works at a high level, then build a small semantic search demo in Python using sentence embeddings and cosine similarity. We’ll also look at why vector databases become useful when you move from a few documents to millions and where vector search fits into systems like RAG. We will go through - Vector search - Semantic search - Cosine similarity - Vector databases - Similarity search - RAG - Machine learning search systems 💻 Code 👉 github repo: https://github.com/afterhoursml/Vector-search If you would like to dive deeper into related concepts - Principal Component Analysis (PCA): https://youtu.be/ejLauPnuK08 - Linear Regression: https://youtu.be/a5b1f77rXsk - Embeddings: https://youtu.be/bUa-hfqhfkA - KNN: https://youtu.be/EujLda19h9E - Why accuracy is not always the best metric: https://youtu.be/TSxtdKB2uH8 If you enjoy this kind of content, consider subscribing and leaving a like & comment. It really helps more curious minds discover the channel. Feel free to share in the comments what your favorite Machine Learning algorithm is. Thanks for watching! See you in the next one. 00:00 Introduction 01:22 Why keyword search is limited 02:15 Intuition 03:02 Python demo 06:10 Cosine similarity 06:49 Why vector databases exist 07:34 Real-world use cases 08:12 Limitations 09:07 How this connects to RAG #machinelearning