1. Welcome to SDS320¶
This course is the final module of the Spatial Data Science minor. It builds on the foundations you developed in earlier SDS modules and gives you space to apply them in a more independent, project-oriented setting.
You have already worked with spatial data, GIS, remote sensing and Python. In SDS320, you will use these foundations to design and implement your own spatial analytics project.
2. What this course is about¶
SDS320 moves from data handling and programming fundamentals toward applied spatial analytics and data-driven modelling.
Selected topics include machine learning and deep learning workflows with geospatial data, object detection, semantic segmentation, instance segmentation, change detection, regression, foundation models and satellite embeddings.
The goal is not to cover every method in full theoretical depth. The goal is to help you understand what different methods can do, when they are useful and how they could support a feasible spatial data science project.
3. Who this course is for¶
This course is for students who have completed earlier SDS modules and are ready to work more independently.
You do not need to remember every Python detail from SDS210. Many students will not have used Python intensively for a few months. The course therefore includes reactivation material and practical examples.
You should, however, be willing to work steadily on your own project, ask questions early and revise your plans when a dataset, method or workflow turns out to be more complicated than expected.
4. What you will develop¶
Throughout the semester, you will develop an independent spatial data science project.
This means that you will:
define a feasible spatial question,
select suitable data and methods,
implement an analytical workflow,
evaluate and interpret your results,
document your decisions,
prepare a reproducible repository,
communicate your findings in presentations and a short report.
The final outcome should show that you can design, implement and explain a reproducible spatial analytics workflow for a complex real-world problem.
5. How to approach the course¶
SDS320 follows a project-driven teaching concept.
Before sessions, you should review the relevant course material and think about how it connects to your project. If applicable, you should also provide a summary of the current state of your project to support discussions and help you make progress. During sessions, you should use the time to clarify questions, discuss decisions, test ideas and get feedback. After sessions, you should continue developing your project while the topic is still fresh.
Our weekly in-person sessions partly comprise lectures on a given topic. However, they are primarily working sessions that require continuous engagement with and progress on your project.
6. What success looks like¶
A successful SDS320 project is one that makes you proud of your achievements. It does not have to be technically perfect. What matters more is that your project motivates you and provides multiple learning opportunities.
Success in this course also means that you work steadily, make thoughtful decisions, document your workflow, respond to feedback and communicate both your results and limitations honestly.
You should be able to explain what you did, why you did it, what worked, what did not work and what you would improve with more time.