In this chapter, you were introduced to the concept of hyperparameter tuning in machine learning. After getting acquainted with the Wine dataset and the Adaptive Boosting classifier, both of which we used for testing throughout this chapter, you were presented with the hyperparameter tuning methods of an exhaustive grid search and its genetic algorithm-driven counterpart. These two methods were then compared using our test scenario. Finally, we tried out a direct genetic algorithm approach, where all the hyperparameters were represented as float values. This approach allowed us to improve on the results of the grid search.
In the next chapter, we will look into the fascinating machine learning models of neural networks and deep learning and apply genetic algorithms to improve their performance.