Summary
In this chapter, you have first seen a general overview of classification and its use cases. You have understood how it is different from anomaly detection, but how it can sometimes still be applied to anomaly detection use cases.
You have learned about five models for online classification of which some are mainly adaptations of offline models, and others are specifically designed for working in an online manner. Both types exist, and it is important to have the tools to benchmark model performance before making a choice for a final model.
The model benchmark that you executed in Python was done in such a way as to find the best model in terms of the accuracy of the model on a test set. You have seen clear differences between the benchmarked models, and this is a great showcase for the importance of model benchmarking.
In the following chapter, you will do the same type of model benchmarking exercise, but this time, you will be focusing on a regression use case, which...
