Implementing RAG using
Large Language Models (LLMs) are only aware of the data they are trained on. They no longer learn once they are deployed in production. One option to mitigate this problem is through Retrieval Augmented Generation or RAG.
In this video, I implement a RAG pipeline using LangChain and ChroDB as the vector store. Towards the end, I chat with my emails to make the LLM answer questions about my emails.
⌚️ ⌚️ ⌚️ TIMESTAMPS ⌚️ ⌚️ ⌚️
0:00 - Intro
0:56 - Getting email (Gmail) Dump
1:58 - About ArXiv
2:45 - Installations
2:55 - LangChain
3:08 - LangChain Hub
4:06 - ChromaDB
5:05 - GPT4All
6:03 - LangChain Imports
7:23 - Preprocessing emails
10:10 - LangChin Text Splitters
11:20 - ChromaDB store
12:39 - LLM model
14:37 - LLM model + RAG
PREVIOUS RELATED VIDEOS
What is RAG (theory) - • What is Retrieval Augmented Generatio...
Fine-tuning LLMs (hands-on) - • Fine-tuning LLMs with PEFT and LoRA -...
Quantizing LLMs (theory) - • Quantization in Deep Learning (LLMs)
Quantizing LLMs (hands-on) - • GGUF quantization of LLMs with llama cpp
MY KEY LINKS
YouTube: / @aibites
Twitter: / ai_bites
Patreon: / ai_bites
Github: https://github.com/ai-bites
WHO AM I?
I am a Machine Learning Researcher/Practitioner who has seen the grind of academia and start-ups equally. I started my career as a software engineer 15 years ago. Because of my love for Mathematics (coupled with a glimmer of luck), I graduated with a Master's in Computer Vision and Robotics in 2016 when the now happening AI revolution had started. Life has changed for the better ever since.
#machinelearning #deeplearning #aibites