workshop-llm

Workshop to create a RAG application using LLM models.

This workshop is developed in Python 🐍 (Jupyter Notebook) and InterSystems IRIS.

The main purpose is to show you the main steps to create a RAG application using an LLM and a vector database.

You can find more in-depth information in https://learning.intersystems.com.

What do you need to install?

Git
Docker (if you are using Windows, make sure you set your Docker installation to use “Linux containers”).
Docker Compose
Visual Studio Code + InterSystems ObjectScript VSCode Extension

Setup

Build the image we will use during the workshop:

Clone the repository:

git clone https://github.com/intersystems-ib/workshop-llm
cd workshop-llm

Build the image:

docker compose build

Run the containers:

docker compose up -d

After running the containers, you should be able to access to:

InterSystems IRIS Management Portal. You can login using superuser / SYS
Jupyter Notebook

Explore RAG applications using Jupyter

Medicine Leaflet examples

You have some medicine leaflets (in spanish) in https://github.com/intersystems-ib/workshop-llm/blob/main/./data.

This example is about creating a RAG Q&A application that can answer questions about those medicine leaflets.

Open Jupyter Notebook, there you can find:

QA-PDF-LLM.ipynb - RAG example using MistralAI LLM
QA-PDF-Local.ipynb - RAG example using a local LLM

alt text

You can test the project step by step or execute it at one time, feel free.

Hoolefoods data model text to SQL

This example is about a company called Holefoods that sells food with some hole on it :)

Using the sales data model of the company, the goal is to create an assistant that can translate natural language questions into valid SQL that answer the question.

In Jupyter Notebook, you will find:

QA-SQL-LLM.ipynb - text to SQL example using OpenAI LLM.

Create other applications

There are some other examples you can try to build and modify in your local environment.

First of all, go to the folder ./python and create a new environment and install some requirements:

# create a local venv environment
# ... for Mac or Linux users
python3 -m venv .venv
# ... for Windows users
python -m venv .venv

# activate venv
# ... for Mac or Linux users
source .venv/bin/activate
# ... for Windows users
./venv/Scripts/Activate.ps1

# install dependencies
pip3 install -r requirements.txt

Create an .env file for storing API keys for OpenAI / MistralAI. They will be used in the applications.

OPENAI_API_KEY="your-api"
MISTRAL_API_KEY="your-api"

Text to SQL service API

You can find a sample Text to SQL based on QA-SQL-LLM.ipynb here.

You can run it like this:

cd python/holefoods_text2sql
fastapi dev main.py

Then open http://127.0.0.1:8000/docs to explore the API and try it out using the web client.

## Streamlit Assistant
There is also a great example of a langchain / streamlit chatbot assitant in https://alejandro-ao.com/how-to-use-streaming-in-langchain-and-streamlit/

You can play with it here as well:

cd python/assitant
streamlit run chatbot.py

Then open http://localhost:8501 and have a look at it.

Are you able to add the logic to reproduce the Medicine Leaflet example in the assitant ?