TitanicML.py 文件源码

def applyRandomForestClassifier(self, train, test):
        #init algorithm 
        RFC = RandomForestClassifier()

        #training target 
        y_train = train[["Survived"]]
        x_train = train[train.columns.difference(["PassengerId","Survived"])]

        #fitting 
        RFC.fit(x_train, y_train)

        result = RFC.predict(test[test.columns.difference(["PassengerId"])])

        self.writeMessage("current training score")
        print RFC.score(x_train, y_train)

        test["Survived"] = result 

        return test