我有一个使用Scikit-Learn的基本决策树分类器：

#Used to determine men from women based on height and shoe sizefrom sklearn import tree#height and shoe sizeX = [[65,9],[67,7],[70,11],[62,6],[60,7],[72,13],[66,10],[67,7.5]]Y=["male","female","male","female","female","male","male","female"]#creating a decision treeclf = tree.DecisionTreeClassifier()#fitting the data to the treeclf.fit(X, Y)#predicting the gender based on a predictionprediction = clf.predict([68,9])#print the predicted genderprint(prediction)

当我运行程序时，它总是输出“male”或“female”，但我如何才能看到预测为男性或女性的概率呢？例如，上面的预测返回“male”，但我如何让它打印出预测为男性的概率呢？

谢谢！

回答：

你可以像下面这样做：

from sklearn import tree#load dataX = [[65,9],[67,7],[70,11],[62,6],[60,7],[72,13],[66,10],[67,7.5]]Y=["male","female","male","female","female","male","male","female"]#build modelclf = tree.DecisionTreeClassifier()#fitclf.fit(X, Y)#predictprediction = clf.predict([[68,9],[66,9]])#probabilitiesprobs = clf.predict_proba([[68,9],[66,9]])#print the predicted genderprint(prediction)print(probs)

理论

clf.predict_proba(X)的结果是：叶子节点中同一类样本的比例，即预测的类概率。

结果的解释：

第一个print返回['male' 'male']，所以数据[[68,9],[66,9]]被预测为males。

第二个print返回：

[[ 0. 1.] [ 0. 1.]]

这意味着数据被预测为男性，并且由第二列的1表示。

要查看类的顺序，请使用：clf.classes_

这将返回：['female', 'male']