You need to deliver unmarked assignments (“Studienleistung”) during the course and at the end a marked student HTML tutorial work as part of your overall course portfolio (“Prüfungsleistung”), see below.
The submissions generally encompass R or R markdown AND the compiled html files. You need to submit your solutions to your personal repository in the GitHub classroom.
The page with the description of this assignment can be found here.
The sessions with the student tutorials will start at 9:15 am.
|Dehn, Hannes||Neural Networks||05.07.2022|
|Kühn, Maira||Neural Networks||05.07.2022|
|Maecker, Luis||Mogel Averaging||12.07.2022|
|Uebe, Leon||Random Forest||21.06.2022|
|Xander, Lasse Finn||Bayesian SDM||05.07.2022|
Use the examples provided here and the video recording (links in Ilias) as orientation.
Plan with 30 minutes minimum up to one hour for each tutorial presentation.
Test your presentation!
Check the technical requirements you need for the live-streaming of your tutorial on your local machine (e.g. audio, video, screen sharing, internet connection) before your presentation.
Some useful data can be found on our course server in the folder “geodata”. The credentials are provided in Ilias.
For creating a seamless raster map of your results, you need a grid system, which you can find on the course server. You may want to use the 50 km x 50 km grid system, which is a compromise between accuracy and processing time.
- The aim of the tutorials is to teach others how your method works and how it is applied. Special emphasis should be given to an appealing, reproducible and self-explanatory HTML page for your method.
- Technically, the result of your modelling workflow should be a map of the species numbers of butterflies in Pakistan. This means each cell of your grid system has as value the sum of the occurring species in this cell. So yes, you need to model all species separately and sum up the occurring species within each cell.
- A possible structure for your tutorial can be found in unit 4 (“SDM workflow”). Do not forget to include the basics for making others understand how your method works (“conceptual underpinning”).
- Make sure that your tutorial is reproducible. If you want to be really professional, include the output of sessionInfo() at the end of your tutorial.
- Start early! This is going to be quite some work. Use the tutorials of former students as orientation.
- The student tutorial will make up 100% of your overall grade. German participants must prepare a tutorial for passing the course.
- If live processing of your script takes too much time (it probably will), use screenshots or intermediate graphical outputs and add them to your HTML page, which should be self-explanatory anyways.
- Describe with text for each processing step why it is done and what for.
- Include references to relevant papers, internet sources, and data.
- Check your spelling.
Good luck! (and start early with your preparations)