In the previous section we worked on ways to preprocess the data and select the most promising features. As we stated, selecting a good set of features is a crucial step to obtain good results. Now we will focus on another important step: selecting the algorithm parameters, known as hyperparameters to distinguish them from the parameters that are adjusted within the machine learning algorithm. Many machine learning algorithms include hyperparameters (from now on we will simply call them parameters) that guide certain aspects of the underlying method and have great impact on the results. In this section we will review some methods to help us obtain the best parameter configuration, a process known as model selection.
We will look ...
No credit card required