VLM Vision Experiments

This repository contains experiments and explorations with NVIDIA's Vision Language Model (VLM) API. Each experiment is documented in its own directory with detailed README files.

Structure

• experiments/: Contains individual experiment directories
• common/: Shared utilities and helper functions
• requirements.txt: Project dependencies

Experiments

1. ID Card Information Extraction: Extract structured information from Spanish ID cards using VLM API

Setup

1. Clone this repository
2. Create a .env file in the project root
3. Add your NVIDIA API key:

   NVIDIA_API_KEY=your_actual_api_key_here
   

4. Install dependencies:

   pip install -r requirements.txt

Security

• NEVER commit your API keys to version control
• Use environment variables to manage sensitive credentials

Contributing

Each new experiment should:

1. Have its own directory under experiments/
2. Include a detailed README.md
3. Document findings and learnings
4. Include sample data (if not sensitive)

License

MIT License - See LICENSE file for details