Skip to content

Example Use Cases#

This toolkit provides a series of tutorial notebooks designed to support geospatial data processing and machine learning tasks on Earth System Data Cubes (ESDCs). Key use cases include:

  1. Land Surface Temperature Prediction: Demonstrates land surface temperature prediction. This serves as an introductory example for ESDC analysis.

  2. Distributed Machine Learning: Showcases efficient preparation and training of large datasets on distributed systems.

  3. Transfer Learning: Illustrates how to reuse pre-trained models for related tasks with limited data.

  4. Cube Insights: Explores the characteristics of data cubes, helping inform preprocessing and modeling decisions.

  5. Gap Filling: Provides techniques to fill missing data in remote sensing datasets using support vector regression (SVR).

  6. ML for Multidimensional Samples with missing values: Demonstrates predictions on multidimensional data, utilizing simpler data imputation methods to address gaps.

Each of these use cases is accompanied by a corresponding Jupyter notebook/Python script, providing step-by-step instructions, code examples, and visualizations to guide users through the entire workflow. Whether you're looking to explore your data, fill gaps, or train machine learning models, this toolkit offers a set of resources to help you achieve your goals.