PDF and Image Data Extraction

This project provides a FastAPI backend for transforming PDFs into images for the purposes of structured data extraction via BAML. It includes a Streamlit frontend for easy interaction and usage.

Key Components

• FastAPI Backend - REST API for processing PDFs into images
• Streamlit Frontend - User-friendly web interface for file uploads and results
• PDF Processing - Converts PDFs to images for analysis
• BAML Integration - Generate a concise summary of the PDF document that was transformed into images

Setup and Installation

Install uv environment

1. Install uv if you don't have it already:

   curl -LsSf https://astral.sh/uv/install.sh | sh

2. Create and activate a virtual environment:

   uv venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install dependencies:

   uv pip install -r requirements.txt

Run FastAPI server

Development Mode

Run the server in development mode with auto-reload:

uv run fastapi dev app/main.py

If you're using environment variables:

uv run dotenv run fastapi dev app/main.py

Production Mode

Run the server in production mode:

uv run fastapi run app/main.py

[Optional] Use Docker for production

If you want to run the FastAPI server in production, you can use Docker.

1. Build the Docker image:

   docker build -t fastapi-app .

2. Run the container:

   docker run -p 8000:8000 fastapi-app

3. For production with environment variables:

   docker run -p 8000:8000 --env-file .env fastapi-app

Running the Streamlit App

1. Make sure the FastAPI server is running first.

2. Run the Streamlit app:

   streamlit run streamlit_app.py

3. Open your browser to the URL provided by Streamlit (usually http://localhost:8501).

4. Upload a PDF or image file, click "Extract Data", and view the results.

5. Extracted images will be saved to the extracted_images directory.

Testing the API Directly

You can test the API endpoint using curl:

curl -X POST -H "Content-Type: multipart/form-data" -F "file=@files/images/invoice.png" http://localhost:8000/extract

Make sure the file path (files/images/invoice.png) points to an existing file on your system.

API Endpoints

The following endpoints are available:

• GET /: Information about the API
• POST /extract: Upload a PDF or image file to extract data using BAML

Add more endpoints that use BAML to extract structured data from the images as necessary.

Demo of the API and Streamlit app

• Run the FastAPI server in development mode
  • uv run fastapi dev app/main.py
• Run the Streamlit app
  • streamlit run streamlit_app.py
• Open your browser to the URL provided by Streamlit (usually http://localhost:8501)
• Upload a PDF or image file, click "Extract Data", and view the summary of the document and a preview of the extracted images (the images are saved to the extracted_images directory)