The notebooks in this folder serve as examples for solutions for the complex exercise, as an LLM would solve it. Most of them would allow to pass as pre-exam requirement even though they lack details:
- Algorithm descriptions could be provided in more detail.
- Some notebooks are using a wrong segmentation algorithm and fail delivering a useful result.
- Installation instructions are commonly not detailed enough. Ideal would be if the notebook was accompanied by a requirements.txt or environment.yml file.
- The extracted feature tables should contain the filename the features were measured in. The notebook notebook_05.ipynb shall be highlighted as it provided the highest segmentation quality as self-tested by the LLM.
In order to make the notebooks run, it was required to update some dependencies:
pip install npe2==0.7.6 pydantic==2.7.4 napari==0.4.19 lxml_html_clean
The used environment is stored in environment.yml.