环境：python 3，scikit-learn 0.18

#coding:utf-8 """ python 3 scikit-learn 0.18 """ from sklearn.model_selection import GridSearchCV from sklearn.model_selection import train_test_split from sklearn.tree import DecisionTreeClassifier from sklearn.metrics import accuracy_score,confusion_matrix,classification_report import input_data import numpy as np mnist = input_data.read_data_sets('mnist/',one_hot=False) x = mnist.train.images y = mnist.train.labels train_data,validation_data,train_labels,validation_labels = train_test_split(x,y,test_size=0.2) #使用GridSearchCV找到最优参数 dtree = DecisionTreeClassifier(random_state=0) #gini ,表示决策树非叶节点划分依据是根 据 gini 指数表示划分的纯度。 #entropy ,用信息增益来衡量 划分的优劣 criterion_options = ['gini','entropy'] splitter_options = ['best','random'] param_griddtree = dict(criterion=criterion_options,splitter=splitter_options) griddtree = GridSearchCV(dtree,param_griddtree,cv=10,scoring='accuracy',verbose=1) griddtree.fit(train_data,train_labels) print('best score is:',str(griddtree.best_score_)) print('best params are :',str(griddtree.best_params_))

结果

耗时4.5min找到最优参数