Hands-On Artificial Intelligence for IoT

Decision trees in scikit

The scikit library provides DecisionTreeRegressor and DecisionTreeClassifier to implement regression and classification. Both can be imported from sklearn.tree. DecisionTreeRegressor is defined as follows:

class sklearn.tree.DecisionTreeRegressor (criterion=’mse’, splitter=’best’, max_depth=None, min_samples_split=2, min_samples_leaf=1, min_weight_fraction_leaf=0.0, max_features=None, random_state=None, max_leaf_nodes=None, min_impurity_decrease=0.0, min_impurity_split=None, presort=False)

The different arguments are as follows:

• criterion: It defines which loss function to use to determine the split. The default value is mean square error (mse). The library supports the use of friedman_mse and mean absolute error ...