I am currently working with extra-trees in the sklearn package but was wondering how the feature importance (function) is calculating the importances of the features. Is it also calculated by the mean decrease Gini or is there another way? Could not find anything in the source code on GitHub.
Extra tree classifier in sklearn used Gini Importance for calculating the feature importance. You can check the following link: http://scikit-learn.org/stable/modules/generated/sklearn.tree.ExtraTreeClassifier.html