Rag-n-Bones

Rag-n-Bones is a web application that uses an input of PDFs to generate systematic reviews utilising Pinecone indexing and OpenAI's GPT models.

Features

• User Authentication
• Prompt Input
• PDF Upload
• Systematic Review Generation
• File Input Access
• Export as PDF
• Quality Check Graphs
• Systematic Review History

Deployment Instructions

Prerequisites

• Latest version of pip is installed
• MySQL installed (database)
• Node.js and npm installed (frontend)

Instructions

1. Clone the repository:

~ git clone https://github.com/pgzqtss/RAG-project.git
~ cd RAG-project

2. Install all dependencies:

Backend

~ cd backend
~ pip install -r requirements.txt

Frontend

~ cd frontend
~ npm install

3. Create a .env file in the root directory using .env.environment as a template.

Environment Variable	Description	Source
PINECONE_API_KEY	API key for using Pinecone as a vector database	Pinecone
OPENAI_API_KEY	API key for use of OpenAI's GPT 3.5 Turbo model	OpenAI API Platform
MYSQL_PASSWORD	Personal password for MySQL database (empty if none)

4. Modify config.py as needed

The config.py file in /backend manages key settings for the application, including:

• Pinecone initialisation for vector storage and retrieval.
• MySQL configuration for database connectivity.
• Embedding and OpenAI model settings for text processing and generation.

5. Start MySQL server

Mac

~ brew services start mysql

Windows

~ net start mysql

Start in cmd through administrator rights

6. Create the MySQL database

~ cd backend
~ mysql -u root -p < schema.sql

7. Run the backend and the frontend at the same time

~ cd backend
~ python3 app.py

(Could run in terminal)

~ cd frontend
~ npm run dev

(Could run in Command Prompt (cmd) for Window)

The Next.js app will run on http://localhost:3000

Testing Instructions

Backend

1. Navigate to the backend directory:

~ cd backend

2. Install the dependencies if not already installed:

~ pip install -r requirements.txt

3. Run the tests using pytest:

~ pytest

This will execute all the tests in the tests directory and provide a summary of the results.