My question are conceptual. We have hyperparameters is a right hyperparameters, can beat you in is the right. I said for the Random Forest metal, the random Forest metal tears. It has some parameters like number of the trees in estimator or maximum number of features. How we can know that. What is the best number of Threes? Okay. What is your what is your approach? Tell me. It is grid search cross-validation and randomized, selfies are so, you should know that. What is, what is this piece of code is all about? Which one is better? What is the difference between clustering? And for example, regression. But is the difference between classification and clustering, the difference is important. What is recommendation?