This repo is part of a series of lectures given at Oslo University Hospital during the fall of 2022 (ML course for Cancer Researchers).
Our aim is to show how different ML models can be tuned, trained and tested using the spam dataset.
Along the way, several Machine Learning principles (e.g. bias-variance trade-off) are demonstrated to gain further insight into each model's behavior.
We used the mlr3 ML framework for model building and evaluation.
A gitbook report is available online.