You started the chapter by understanding predictive analysis with regression and explored the concepts of training, testing, and evaluating a regression model. You then proceeded to carry on building experiments with different regression models, such as linear regression, decision forest, neural network, and boosted decision trees inside ML Studio. You learned how to score and evaluate a model after training. You also learned how to optimize different parameters for a learning algorithm with the Sweep Parameters module. The No Free Lunch theorem teaches us not to rely on any particular algorithm for every kind of problem, so in ML Studio you should train and evaluate the performance of different models before finalizing a single one.

In the next chapter, you will explore another kind of unsupervised learning called classification and you will explore the different algorithms available with ML Studio.