reported during quality assurance:
Lesson 2 data exercise with synthetic dataset: the learners were confused about the fact that both the exercise and the dataset column "label" indicates 2 classes, while the scatterplot clearly reveals 4 classes (especially since the distribution of binary class labels was seemingly random). So this should be fixed.