GeoAI in the Cloud: From Data Formats to Foundation Models

Welcome to the EarthSense 2025 EOFM tutorial 🥳! This Jupyter book 📖 contains notebooks on how to apply 🧱 Foundation Models to 🛰️ Earth Observation data.

📚 Overview of tutorials¶

1. Introduction to Cloud Native Data Formats

2. Introduction to EO Foundation Models

3. Applying Foundation Models to Segmentation downstream task

Each tutorial is rendered on this website for easy viewing 👀, but some of them are Jupyter notebooks designed to be ran interactively 💫. See the instructions below on how you can start running the tutorials in no time! 🚀

🌠 Setting up your environment¶

To run these notebooks in an interactive Jupyter session online, 🖱️ click on the button below to launch on Google Colaboratory.

Notebooks

• Exploring Cloud Native Geospatial Data Formats with ERA5 Reanalysis Data

• Fine-tuning Clay Foundation Model for Land Cover Segmentation

💻 Creating a local environment for running tutorials¶

If you prefer to run the tutorials locally, follow these simple steps:

1. Ensure you have uv installed.

2. Clone the repository and set up the environment:

   git clone https://github.com/developmentseed/earthsense25tutorial.git
   cd earthsense25tutorial
   uv sync

3. Launch Jupyter Lab:

   uv run jupyter lab

You’re now ready to run the tutorials! 🎉