Random Forests

As briefly mentioned earlier, random forests are ensembles of decision trees that can be used to solve classification or regression problems. Random forests use a small portion of the data to fit each tree, so they can handle very large datasets, and they are less prone to the "curse of dimensionality" relative to other algorithms. The curse of dimensionality is a situation in which an abundance of features in the data diminishes the performance of the model. Predictions of the random forest are then determined by combining the predictions of each tree. Like SVM, random forests are a black box with inputs and outputs which cannot be interpreted.

In the upcoming exercises and activities, we will tune and fit a random forest regressor using grid search to predict the temperature in Celsius. Then, we will evaluate the performance of the model.

Exercise 32: Preparing Data for a Random Forest Regressor

First, we will prepare the data for the random forest regressor with...