The NEODAAS team successfully delivered a comprehensive half-day Introduction to Machine Learning for Earth Observation training workshop as part the Machine Learning for Earth Observation Conference (ML4EO.org) hosted at the University of Exeter, introducing 22 participants to the practical applications of machine learning in satellite data analysis. The workshop was specifically designed for Earth Observation users with limited machine learning experience but basic Python knowledge, addressing a growing need in the EO community to understand and apply AI techniques to satellite data analysis.
Participants gained practical experience through a series of Jupyter notebooks, working through applications ranging from simple random forest models to advanced neural network techniques. The training leveraged NEODAAS’s MAGEO (Massive Graphics Processing Unit Cluster for Earth Observation), allowing participants to compare CPU and GPU processing performance firsthand. The MAGEO cluster provides researchers with access to high-performance GPU computing specifically optimized for Earth Observation applications, enabling faster processing of machine learning workflows that would be impractical on standard computing infrastructure.
One participant’s feedback about the course was “I found this course really useful for demonstrating ML methods by linking it with existing climate analysis tools I am already familiar with”.
The success of this workshop demonstrates the growing appetite for accessible machine learning training in the Earth Observation community. NEODAAS continues to bridge the gap between traditional EO analysis and modern AI techniques, making advanced computational tools accessible to a broader range of researchers and practitioners.
For more information about NEODAAS training opportunities and resources, visit NEODAAS.ac.uk or follow us on LinkedIn https://www.linkedin.com/company/neodaas