Radiant MLHub hosts open ML training datasets generated by Radiant Earth Foundation, partners, and community. Designed to encourage widespread data collaboration, Radiant MLHub allows anyone to access, store, register, and share open training datasets for high-quality Earth observations.
Community of Practice
Radiant MLHub facilitates an open community commons for geospatial training data, machine learning models, and standards to encourage collaboration and share information. Find out how you can get involved.
API and Python Client
The Python client allows users to search and download geospatial training data on Radiant MLHub without managing API requests. Or users may apply MLHub with other scripting languages using our REST API. View the Python quick-start and Jupyter Notebooks in the Documentation section.
Geospatial Machine Learning Model Catalog
Radiant Earth is developing the Geospatial Machine Learning Model Catalog (GMLMC) specification. The GMLMC will empower users to discover and access existing repositories of ML models for various geospatial applications. View our ML Models specification.
All Radiant MLHub geospatial training data collections are stored using a SpatioTemporal Asset Catalog (STAC) compliant catalogs, and exposed through a common API. Learn more about STAC.