Nanonets OCR2 Cookbooks

A comprehensive collection of examples and configurations for getting the best results from the Nanonets-OCR2 model. This repository provides practical implementations for various OCR use cases including financial documents, complex tables, and multilingual content.

📚 Documentation

Cookbooks & Examples

• Image to Markdown Converter: Comprehensive notebook showcasing the best practices for using Nanonets-OCR2-3B model across various document types including:
  • Bank statements and financial documents
  • Complex tables and structured data
  • Photos captured from mobile devices
  • Multilingual images and documents

🛠️ Installation

Prerequisites

• Python 3.11
• CUDA-compatible GPU (recommended for optimal performance)
• uv package manager

Setup Instructions

1. Install uv package manager:

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

2. Create virtual environment and install dependencies:

   # Create virtual environment with Python 3.11
   uv venv --python=3.11
   
   # Activate the virtual environment
   source .venv/bin/activate
   
   # Install dependencies
   uv pip install -r requirements.txt

3. Start Jupyter Lab:

   jupyter lab

📦 Dependencies

This project uses the following key dependencies:

• PyTorch: Deep learning framework with CUDA support
• Transformers: Hugging Face transformers library for model inference
• Jupyter: Interactive notebook environment
• PDF2Image: PDF document processing
• Accelerate: Model acceleration and optimization

🎯 Quick Start

1. Clone this repository
2. Follow the installation steps above
3. Open the Jupyter notebook in Nanonets-OCR2-Cookbook/image2md.ipynb
4. Run the cells to see examples of OCR processing

🤝 Contributing

Contributions are welcome! Please feel free to submit pull requests or open issues for bugs and feature requests.