Initial Release
Workshop to create a RAG application to use LLM models. This workshop is developed in Python using Jupyter Notebook connected using the official libraries to IRIS.
You can find more in-depth information in https://learning.intersystems.com.
Build the image we will use during the workshop:
$ git clone https://github.com/intersystems-ib/workshop-llm
$ cd workshop-llm
$ docker-compose build
The main purpose of this example is to identify the main steps to create a RAG application using MISTRAL as LLM and IRIS as vector database to save and search the specific context.
There is a known issue related to the permission request from Docker Desktop to access to the folders of the project, this permission has to be granted before to launch docker-compose up -d. To allow the file sharing in Docker Desktop you have to open settings option, select Resources and File Sharing, from that screen you have to include the path to the project, you can see here an example:
If you don’t share this folder previously PostgreSQL database won’t be initialized and the project will fail.
docker-compose build
docker-compose up -d
Automatically an IRIS instance, will be deployed, a Jupyter Notebook is deployed under (http://localhost:8888) too.
superuser
/ SYS
account.Test
This project is devolped in Python using Jupyter Notebook, you can access to it from here and open LLMTests.ipnyb file.
You can test the project step by step or execute it at one time, feel free.