Decision Tree & Random Forrest

On this page are information on how to use decision trees and random forests.

Decision Tree

This implementation follows the typical binary decision tree algorithm and it is more modular. The following modules are available:

• Feature Selection
• Impurity Measure
• Leaf Function
• Split Algorithm

Of these only split algorithms and impurity measure are necessary for training. Leaf functions will be require when making predictions.

Random Forests

Random Forests are simple collections of decision trees.