LLM Bootcamp - Day 2 Highlights

Igniting Innovation: Day 2 Recap from LLM Bootcamp! 🚀
Day 2 at the LLM Bootcamp was a powerhouse of learning and hands-on experience, pushing the boundaries of what's possible with Large Language Models. Here's what our participants dived into:
Introduction to Vector Databases with Zain Hasan:
• Theory Session: Zain captivated the audience with a deep dive into the fundamental concepts of vector databases. He covered critical topics including indexing techniques such as Hierarchical Navigable Small World (HNSW), Product Quantization (PQ), and Locality-Sensitive Hashing (LSH).
• Retrieval Methods: Participants learned about various retrieval techniques like cosine similarity, Euclidean distance, and dot product. These methods are essential for applications in video and image recognition, natural language processing (NLP), recommender systems, and particularly Retrieval-Augmented Generation (RAG).
• Hands-On Lab: The theory was followed by a hands-on lab where attendees prepared datasets with vectors, wrote data schemas for a vector database using Redis, stored data, and created vector search indexes. They performed complex queries on a vector database, including tag filters, numeric filters, text filters, geographic filters, combining filters, and range queries. The session also included using built-in text embedding vectorizers like OpenAI and HuggingFace.
🌟Feeling inspired? Register now and join us on this transformative journey into the world of AI: hubs.la/Q02DttMK0