This project focuses on automating the verification of PDF documents against specific requirements using advanced AI capabilities. The process begins by converting each page of the PDF into an image format. These images are then passed, along with the prompt, to the OpenAI GPT-4 model through the API for evaluation. The final output is a JSON Object.
This automated solution streamlines the process of document compliance checks, making it ideal for industries where accuracy and consistency in documentation are critical.
Clone the repository:
git clone https://github.com/devAdityaa/automee_Assignment.git
Go to the project directory:
cd automee_Assignment
Install the requirements:
npm install
This script uses pdf2pic which needs the following packages in your system: GraphicsMagick (GM).
- For Mac Environment
brew install graphicsmagick
- For Linux:
sudo apt-get install graphicsmagick
A sample .env file is already provided, users can remove the .sample
from the file name to make it a .env file.
All the env variables are given inside of it.
To run the project use:
node index.js