Understanding model parameters
In your data science workload, as you define your features, you determine which parameters should be leveraged by your model. However, depending on the algorithm selected, you can control the training behavior by altering the parameters of the model itself – this is known as hyperparameter tuning. Using hyperparameter tuning, we can explore a variety of model parameters to identify the best model parameters to establish the best model result. To evaluate the model results, a primary metric is selected. A primary metric is defined as the key metric for evaluating the model. Every time a hyperparameter is changed, the primary metric will either go up or down in value and based on the primary metric, that will yield a better or worse model.
In this chapter, we will create a logistic regression model by leveraging sklearn’s implementation of logistic regression with an sklearn pipeline. For a logistic regression model, there are several model...
