控制决策树的深度

控制决策树的深度

min_samples_split=2,
叶子节点需要的最下样本数
min_samples_leaf=1,
min_weight_fraction_leaf=0,
最大叶子节点数
max_leaf_nodes=None,
执行分裂所需最小不纯度减少量
min_impurity_decrease=0)

# 控制决策树的规模  树的深度
from sklearn.tree import DecisionTreeClassifier
from sklearn import datasets
# 加载数据
iris = datasets.load_iris()
features = iris.data
target = iris.target
# 
# 创建决策树分类器对象
decisiontree = DecisionTreeClassifier(random_state=0,
                                     max_depth=None,
#                                       节点分裂前，节点上最小样本数
                                     min_samples_split=2,
#                                       叶子节点需要的最下样本数
                                     min_samples_leaf=1,
                                     min_weight_fraction_leaf=0,
# 最大叶子节点数
                                     max_leaf_nodes=None,
#                                       执行分裂所需最小不纯度减少量
                                     min_impurity_decrease=0)
# 训练模型
model = decisiontree.fit(features, target)