Chat with SQL and Tabular Databases using LLM Agents (DON'T USE RAG!)

In this video, together we will go through all the steps necessary to design a ChatBot APP to interact with SQL and Tabular Databases using natural language, SQL LLM agents, and GPT 3.5. We will design a Chatbot that can:
1. Chat with SQL DB that we create from SQL files
2. Chat with SQL DB that we create from CSV and XLSX files
3. Chat with SQL DB that we create by uploading documents while using the chatbot
4. RAG with Tabular data

Moreover, in this video, I will show you why RAG is not a good option for interacting with your databases. The code is available on the Github repository.

🚀 GitHub Repositories:
Advanced Q&A and RAG series: https://github.com/Farzad-R/Advanced-...
LLM-Zero-To-Hundred Series: https://github.com/Farzad-R/LLM-Zero-...

00:00 Intro
00:41 Roadmap (Q&A vs RAG)
05:59 Resources of the first project
09:35 Project schema walk-through
13:19 Test your GPT and Embedding models (Notebook)
14:44 How to load environment variables (.env file)
17:26 Step 1.1: Create the SQL database from .sqlfile
19:00 Step 1.2: Create the SQL database from CSV and XLSX files
20:22 Step 1.3: Create the VectorDB from CSV and XLSX files
21:25 Step 2: Test your SQL database (Notebook)
22:15 Step 3: Step-by-step guide for Q&A with the SQL database created from SQL file (Notebook)
31:15 Step 4: Q&A with the SQL database created from CSV and XLSX files (Notebook)
36:15 Step 5: RAG with SQL databases and Tabular Data (Notebook)
42:34 ChatBot GUI brief backend walk-through
47:00 UI walk-through
48:25 Demo: Q&A with SQL DB created from .sql file
50:36 Demo: Q&A with SQL DB created from CSV and XLSX files
53:03 Demo: RAG with VectorDB created from CSV and XLSX files
54:45 Demo: Q&A with SQL DB created from uploaded CSV and XLSX files
56:55 Keynotes

Langchain SQL Agent: https://python.langchain.com/docs/use...

Frameworks: #langchain , #openai, gradio, SQLite
#chatbot #rag #llm #agent #python #gpt