In the current exercise a neural network is trained on the Diabetes data set. To educate the students on which general use cases exist and which training frameworks would typically be used, I propose to modify this chapter as follows:
Following the discussion of the machine learning concepts, introduce the concepts of structured and unstructured data with practical examples.
Give a short overview of the most common training frameworks and typical use cases with respect to structured and unstructured data problems, e.g. scikit-learn, xgboost, lightgbm, catboost, tensorflow, pytorch.
Change the model trained in the existing Diabetes problem to a scikit-learn type model.
Add another exercise for working on an unstructured data problem, for which a deep learning model with PyTorch is used.
noelo
commented
3 days ago
In the current exercise a neural network is trained on the Diabetes data set. To educate the students on which general use cases exist and which training frameworks would typically be used, I propose to modify this chapter as follows: