Sindri, named after the God of War character who always has your back, is an advanced privacy policy analysis tool. It leverages state-of-the-art machine learning techniques to comprehensively annotate privacy policies for data collection and opt-out measures.
- Fine-tuned Meta's Llama-3 8B LLM for text classification
- Comprehensive privacy policy annotation
- Robust text pre-processing pipelines
- CLI interface for easy interaction
- Python
- PyTorch
- Scikit-Learn
- CUDA
- Pandas
- HuggingFace Transformers
- Deepset's Haystack
- SpaCy
- Docker
- AWS Inferentia2 (for deployment)
- Python 3.8+
- CUDA-compatible GPU (for training)
- Docker (for deployment)
-
Clone the repository:
git clone https://github.com/your-username/sindri-privacy-analyzer.git cd sindri-privacy-analyzer -
Create a virtual environment:
python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate` -
Install the required packages:
pip install -r requirements.txt
To use Sindri, run the main CLI program:
python main.py
Follow the prompts in the CLI to analyze privacy policies. The program will guide you through the process of inputting a privacy policy and receiving the analysis results.
Sindri is designed to be deployed using Docker and AWS Inferentia2.
This README provides a comprehensive overview of your project, including its features, technologies used, installation process, and usage instructions. You may want to add or modify sections as needed, especially regarding specific installation requirements, detailed usage instructions, or any other project-specific information.
Remember to add a license to your project if you haven't already, and update the repository URL in the installation instructions to match your actual GitHub repository.
Is there anything else you'd like me to add or modify in this README?