This repository is the result of a project conducted at the University of Münster in the Geoinformatics bachelors program as part of the course "Geosoftware II."
This project aims to develop a user-friendly platform that facilitates access to pre-trained Machine Learning models for Earth Observation (EO) data. The platform allows users to search, upload, and manage models and their metadata. It implements a STAC-compliant structure and provides a REST API for seamless integration into existing workflows. The system is built with React, Node.js, and PostgreSQL, utilizing Docker for deployment. Future enhancements could include expanded filtering options, full administrative functionality, and the ability for users to edit or delete their models.
- User-Friendly Interface: Easy navigation for model management.
- STAC-Compliant Structure: Standardized metadata for EO data.
- REST API Integration: Seamless connection to external systems.
- Dockerized Deployment: Simplified installation and scaling.
- Search & Upload: Quickly find and add ML models.
First you need to have Docker installed and running.
-
Clone the Repository
git clone https://github.com/awiechma/CatX.git
-
Navigate into the directory
cd CatX -
Run the App using docker
docker compose up --build
If you wish to use demo data:
-
Navigate into the directory
cd ./demo_data -
Run the script in the terminal:
bash insert_demo.sh
If Step 2 does not work:
- Navigate into the directory in git bash:
cd CatX/demo_data - Run the script using git bash:
insert_demo.sh
After starting the app, visit http://localhost:5173 in your browser. Use the intuitive interface to explore and upload ML models.
Distributed under the MIT License. See LICENSE.txt for more information.