Decision Tree Hyperparameters
Hyperparameter Description Possible Values
criterion Function to measure the quality of a split classification: 'gini', 'entropy'; regression: 'squ
splitter Strategy used to choose the split at each node 'best', 'random'
max_depth Maximum depth of the tree int or None
min_samples_split Minimum number of samples required to split an internal
int, float
node
min_samples_leaf Minimum number of samples required to be at a leafint,
node
float
max_features Number of features to consider when looking for theint,
best
float,
split'auto', 'sqrt', 'log2', None
random_state Controls the randomness of the estimator int, RandomState instance, None
max_leaf_nodes Maximum number of leaf nodes in the tree int, None
min_impurity_decrease A node will be split if this split induces a decrease infloat
impurity greater than or equal to this value