Summary
In this chapter and the previous one, we explored the basics of ML.NET using AutoML to perform classification and regression problems. We saw how AutoML lets us define the problem we want to solve and the data to solve it, then goes to work evaluating different candidate models until the best model is found within the time given to evaluate models.
While classification focuses on sorting data into different categories, regression’s focus is on predicting a single numerical value – in the case of this chapter, that was the market value of a player in euros.
Like classification, regression models involve metrics. These metrics focus on the average amount of error (sometimes referred to as loss) that is present in the model’s predictions.
We saw how MAE is a simple measurement used to show how close our model usually gets, while RMSE can give us a similar measurement that is more profoundly affected by large errors in the model’s predictions...
