机器学习笔记－－分类 决策树－－人体运动状态信息评级

1.数据集

参考https://blog.csdn.net/qq_39011567/article/details/102732543对数据集的介绍

2.决策树算法分析

参考https://blog.csdn.net/qq_20106375/article/details/94158472对决策树算法的介绍

3.实验过程

1、使用算法：DecisionTreeClassifier分类算法
2、实现过程：
1、建立工程，导入sklearn相关包

import numpy as np

from sklearn.tree import DecisionTreeClassifier

from sklearn.model_selection import train_test_split 

from sklearn.metrics import classification_report

2、加载数据，划分训练集和测试集

    X_train,Y_train=load_datasets(featurePaths[:4], labelPaths[:4])#前四个文件作为训练集

    X_test,Y_test=load_datasets(featurePaths[4:], labelPaths[4:])#后两个文件作为测试集  

    X_train,x_,Y_train,y_=train_test_split(X_train,Y_train,test_size=0.0)#将训练集打乱

3、创建GaussianNB算法实例GNB，并进行训练，获得标签

    dt = DecisionTreeClassifier()

    dt.fit(X_train,Y_train)

4、调用dt.score()方法查看分类准确率，调用classification_report(Y_test,Y_predict)查看分类效果

    accrucy=dt.score(X_test,Y_test)

    classification_report=classification_report(Y_test,Y_predict)

4.实验代码

 

# -*- coding: utf-8 -*-
"""
Created on Mon Oct 28 17:16:41 2019

@author: zhang
"""


 

import pandas as pd

import numpy as np  

 

from sklearn.preprocessing import Imputer

from sklearn.model_selection import train_test_split 

from sklearn.tree import DecisionTreeClassifier

from sklearn.metrics import classification_report

 

def load_datasets(feature_paths, label_paths):

    feature = np.ndarray(shape=(0,41))

    label = np.ndarray(shape=(0,1))

    for file in feature_paths:

        df = pd.read_table(file, delimiter=',', na_values='?', header=None)

        imp = Imputer(missing_values='NaN', strategy='mean', axis=0)

        imp.fit(df)

        df = imp.transform(df)

        feature = np.concatenate((feature, df))

     

    for file in label_paths:

        df = pd.read_table(file, header=None)

        label = np.concatenate((label, df))

         

    label = np.ravel(label)

    return feature, label

 

if __name__=="__main__":

    featurePaths = [r'C:\Users\l\Desktop\ML\naive_bayes\A.feature',r'C:\Users\l\Desktop\ML\naive_bayes\B.feature',r'C:\Users\l\Desktop\ML\naive_bayes\C.feature',r'C:\Users\l\Desktop\ML\naive_bayes\D.feature',r'C:\Users\l\Desktop\ML\naive_bayes\E.feature']

    labelPaths = [r'C:\Users\l\Desktop\ML\naive_bayes\A.label',r'C:\Users\l\Desktop\ML\naive_bayes\B.label',r'C:\Users\l\Desktop\ML\naive_bayes\C.label',r'C:\Users\l\Desktop\ML\naive_bayes\D.label',r'C:\Users\l\Desktop\ML\naive_bayes\E.label']

    X_train,Y_train=load_datasets(featurePaths[:4], labelPaths[:4])#前四个文件作为训练集

    X_test,Y_test=load_datasets(featurePaths[4:], labelPaths[4:])#后两个文件作为测试集

    X_train,x_,Y_train,y_=train_test_split(X_train,Y_train,test_size=0.0)#将训练集打乱

    

    dt = DecisionTreeClassifier()

    dt.fit(X_train,Y_train)

    Y_predict=dt.predict(X_test)

    accrucy=dt.score(X_test,Y_test)

    classification_report=classification_report(Y_test,Y_predict)

    

    print(accrucy)

    print(classification_report)