A guide on using the Spatiotemporal Asset Catalog (STAC) Machine Learning model extension, developed during the 2023 Gaia Pod Week STAC ML Extension hackathon.
Goals:
- Create an end-to-end tutorial on how to use the STAC ml-model extension
- Discover what limitations exist, and what can be improved
To help out with development, start by cloning this repo-url
git clone <repo-url>
Then, create a Python virtual environment to install the dependencies in. We recommend using mamba to initialize the virtual environment, but you can use others too. The virtual environment should be created with Python and pip installed.
cd stacmlguide
mamba create --name stacml python=3.11 pip=23.2.1
Activate the virtual environment first.
mamba activate stacml
Install the dependencies from the pyproject.toml file using pip. This will
include several Python packages and development related dependencies like
JupyterLab.
pip install --editable ".[dev]"
Finally, double-check that the libraries have been installed.
mamba list
- STAC ML Model Extension
- Blog posts
- https://medium.com/radiant-earth-insights/geospatial-models-now-available-in-radiant-mlhub-a41eb795d7d7
- https://medium.com/radiant-earth-insights/discoverable-and-reusable-ml-workflows-for-earth-observation-part-1-e198507b5eaa
- https://medium.com/radiant-earth-insights/discoverable-and-reusable-ml-workflows-for-earth-observation-part-2-ebe2b4812d5a
- Sprint format
- Mentored sprints - https://mentored-sprints.netlify.app/organisers/01-index