vision-parse 0.1.0

Extract content from PDF documents using Vision LLMs in markdown format

Vision Parse

🚀 Transform PDF documents into beautifully formatted markdown using state-of-the-art Vision Language Models - all with just a few lines of code!

🎯 Introduction

Vision Parse harnesses the power of Vision Language Models to revolutionize document processing:

• 📝 Smart Content Extraction: Intelligently identifies and extracts text, tables, and images with high precision
• ✨ Markdown Magic: Converts complex document layouts into clean, well-structured markdown
• 🎨 Format Preservation: Maintains document hierarchy, styling, and visual elements
• 🤖 AI-Powered Analysis: Leverages cutting-edge vision models for superior accuracy
• 🔄 Batch Processing: Handle multi-page documents effortlessly

🚀 Getting Started

Prerequisites

• 🐍 Python >= 3.8
• 🖥️ Ollama (for local model hosting)

Installation

Install the package using pip:

pip install vision-parse

Setting up Ollama

1. Install Ollama based on your operating system:

   Linux:

   curl -fsSL https://ollama.com/install.sh | sh
   

   MacOS:

   brew install ollama
   

   Windows: Download and install from Ollama Website

2. Pull and start the Ollama server:

   ollama pull llama3.2-vision:11b
   ollama serve
   

3. Verify server status:

   curl http://localhost:11434/api/version
   

⌛️ Usage

Basic Example

from vision_parse import VisionParser

# Initialize parser
parser = VisionParser(
    model_name="llama3.2-vision:11b",
    temperature=0.7,
    top_p=0.7
)

# Convert PDF to markdown
pdf_path = "path/to/your/document.pdf"
markdown_pages = parser.convert_pdf(pdf_path)

# Process results
for i, page_content in enumerate(markdown_pages, 1):
    print(f"\n--- Page {i} ---\n{page_content}")

Custom Configuration

from vision_parse import VisionParser, PDFPageConfig

# Configure PDF processing settings
page_config = PDFPageConfig(
    dpi=400,
    color_space="RGB",
    include_annotations=True,
    preserve_transparency=False
)

# Initialize parser with custom config
parser = VisionParser(
    model_name="llama3.2-vision:11b",
    temperature=0.7,
    top_p=0.7,
    page_config=page_config
)

# Convert PDF to markdown
pdf_path = "path/to/your/document.pdf"
markdown_pages = parser.convert_pdf(pdf_path)

🛠️ Development

Setting Up Development Environment

1. Clone and Setup:

   # Clone the repository
   git clone https://github.com/iamarunbrahma/vision-parse.git
   cd vision-parse
   

2. Install Dependencies:

   # Install uv (Mac and Linux)
   curl -LsSf https://astral.sh/uv/install.sh | sh
   
   # Install uv (Windows)
   powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
   
   # Install dependencies
   uv sync --all-extras && source .venv/bin/activate
   

3. Quality Checks:

   # Run test suite
   make test
   
   # Code quality
   make lint    # Run code linting
   make format  # Format code
   

📄 License

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