Medical RAG

this is an implemneation of RAG archiecutre for answering questions related to the medical field

Requirments

python 3.8 or above the version used was 3.12.3 using WSL on windows or using linux natively is completly up to you i prefer linux at this point as it is easier in some cases

to create a virtual enviroment use the following command

python -m venv venv

the second venv is the name of the virtual enviroment you can change it to any name you like to activate the virtual enviroment use the following command

source venv/bin/activate

on windows use the following command

venv\Scripts\activate

after activating the virtual enviroment you need to install the requirments using the following command

pip install -r requirements.txt

running the app

if you want to run the app use the following command

uvicorn src.app.main:app --reload --host 0.0.0.0 --port 5000

this will run the app on localhost port 8000 you can change the port to any port you like also you can change localhost to 0.0.0.0 to make it accessible from other devices on the same network remove it to run the app in development and -- reload is used to reload the app automatically when you make changes to the code to make sure everything is working you can access the following endpoint you can just use

uvicorn src.app.main:app

testing the app

to test the apis you can use pytest and use the file test_api.py to test the apis as you would create more apis you can add more tests to the test_api.py file to test the new apis to run the tests use the following command

pytest

docker and MongoDB

in order to store the chunks we generated from the documents we need to store it in a database beacuse of this we need to use mongoDB , just you need to famalirize yourself just a bit with docker and docker-compose first thing you need is to install docker this is the docker url

Docker Desktop for windows

how to install docker on linux

afer verifying that docker works on your system go the directory where the docker file presents

cd docker , docker compose up --watch

a quick guide on docker compose and what the commands mean

your mongodb will be running on port 27017 we can download studio 3t to view your database studio 3t