The combination of School and ML suggests an educational angle on machine learning. I am most interested in examples that make the connection between data and a result understandable without hiding the limitations. That is what I associate with [School ML machine learning](https://school-ml.edu.pl). A demonstration can invite questions about what the model has learned and what the example does not show. I like explanations that leave room for those questions, because a neat result should not be mistaken for a complete account of how the tool works.