Welcome to the STAR (Standards Technical Assistance Resource) project, developed as part of NASA's SpaceApps Hackathon. STAR aims to revolutionize how technical standards are accessed and understood by leveraging AI and natural language processing.
The STAR project is designed to streamline the process of accessing and comprehending technical standards documents, particularly those related to NASA. We have developed an AI-powered assistant that can assist users in understanding, navigating, and making sense of complex standard documents.
- AI-Powered Recommendations: The AI assistant recommends section numbers, language improvements, and more within standards documents.
- Web Scraping: We scrape NASA's standards website to collect a library of standards documents for analysis.
- FastAPI Web Application: The AI assistant is accessible through a user-friendly web interface.
Our project leverages the following technologies and libraries:
- HTML/CSS/JS for the frontend
- Python
- FastAPI
- PyTorch
- Hugging Face Transformers
- Ngrok
- BeautifulSoup
To set up the STAR project locally, follow these steps:
- Clone the repository:
git clone https://github.com/elucidator8918/Nasa-Space-Apps.git - Navigate to the project directory:
cd Nasa-Space-Apps - Install the required Python packages:
pip install -r requirements.txt - Set up Ngrok for exposing your local server to the internet.
To run the project and access the AI assistant, follow these steps:
- Run the FastAPI server:
uvicorn main:app --reload - Access the AI assistant through your web browser by visiting the provided Ngrok URL.
The following team members developed this project:
This project is licensed under the CC0-1.0 License.