Invoice Extractor

A Python application for extracting structured invoice data from PDF files using multiple AI models (DeepSeek, OpenAI, Mistral) with OCR and exporting results to Excel.

Features

• PDF Text Extraction: Extracts text using PyPDF2, PyMuPDF, and OCR (Tesseract via pdf2image).
• Multi-Model Support: Extracts structured invoice data using DeepSeek, OpenAI, or Mistral APIs.
• Chunk Processing: Splits large PDFs into chunks for efficient processing.
• Excel Export: Exports data to formatted Excel files using pandas and openpyxl.
• Duplicate Handling: Removes duplicate invoices based on vendor name and invoice number.

Prerequisites

• Python 3.8+
• Tesseract OCR installed (Tesseract Installation)
• API keys for DeepSeek, OpenAI, and/or Mistral (set in .env file)
• Poetry for dependency management

Installation

1. Clone the repository:

   git clone https://github.com/magicjohnson/invoice-extractor.git
   cd invoice-extractor

2. Install Poetry:

   pip install poetry

3. Install dependencies using Poetry:

   poetry install

4. Create a .env file in the project root and add your API keys:

   DEEPSEEK_API_KEY=your_deepseek_keyOPENAI_API_KEY=your_openai_keyMISTRAL_API_KEY=your_mistral_key

Dependencies

Managed via Poetry (pyproject.toml):

• PyPDF2
• pdf2image
• pytesseract
• PyMuPDF
• pandas
• openpyxl
• requests
• python-dotenv

Usage

1. Place your PDF invoice file (e.g., invoices_example.pdf) in the project directory.
2. Activate the Poetry virtual environment:

   poetry shell

3. Run the desired extraction script:

   python extract_invoices_deepseek.py # For DeepSeek
   python extract_invoices_openai.py # For OpenAI
   python extract_invoices_mistral.py # For Mistral

4. The script will:
   • Extract text from the PDF using OCR.
   • Process text with the chosen AI model to extract structured invoice data.
   • Export results to extracted_invoice_data.xlsx.
   • Print extracted data to the console.

Example Output

The extracted data is saved in extracted_invoice_data.xlsx with columns:

• Vendor Name
• Invoice Number
• Invoice Date
• Due Date
• PO Number
• Total Amount
• Description
• Bill To
• Payment Terms
• Payment Instructions

Console output example:

Extracted Invoice Data:
Invoice 1:
Vendor Name: Example Vendor
Invoice Number: INV12345
Invoice Date: 2025-09-01
Due Date: 2025-10-01
PO Number: PO67890
Total Amount: $1000.00
Description: Consulting Services
Bill To: Oaks at Creekside
Payment Terms: Net 30
Payment Instructions: Wire transfer to account XYZ

Project Structure

invoice-extractor/
├── extract_invoices_deepseek.py # DeepSeek-based extraction script
├── extract_invoices_openai.py # OpenAI-based extraction script
├── extract_invoices_mistral.py # Mistral-based extraction script
├── invoices_example.pdf # Sample invoice PDF
├── pyproject.toml # Poetry configuration
├── poetry.lock # Poetry lock file
├── .env # Environment variables (not tracked)
└── README.md # This file

Contributing

1. Fork the repository.
2. Create a feature branch (git checkout -b feature/your-feature).
3. Commit your changes (git commit -m 'Add your feature').
4. Push to the branch (git push origin feature/your-feature).
5. Open a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• DeepSeek, OpenAI, and Mistral APIs for structured data extraction
• Tesseract OCR for text extraction
• PyMuPDF and PyPDF2 for PDF processing
• Poetry for dependency management