机器学习实战-kNN笔记

最新推荐文章于 2022-04-17 19:44:32 发布
最新推荐文章于 2022-04-17 19:44:32 发布
阅读量355 收藏
点赞数
分类专栏： python 机器学习 文章标签： 机器学习 kNN python
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/Eclipsesy/article/details/77472516
版权 
import numpy as np
import operator
'''
《机器学习实战》kNN分类算法及注释
'''
'\n《机器学习实战》kNN分类算法及注释\n'

#创建数据集
def createDataSet():
    group = [[1.0,1.1], [0.9,1.0], [0.1,0.2], [0.2,0.3]]
    labels = ['A', 'A', 'B', 'B']
    return group,labels
#numpytile函数,重复[1,2] 3行4列
tile_result = np.tile([1,2],(3,4))
print(tile_result)
[[1 2 1 2 1 2 1 2]
 [1 2 1 2 1 2 1 2]
 [1 2 1 2 1 2 1 2]]

shape_result = tile_result.shape
print(shape_result)
print(shape_result[0])
print(shape_result[1])
(3, 8)
3
8

#定义kNN分类函数
def classify0(inX, dataSet, labels, k):
    #inX代表输入的数据坐标，DataSet:n个m维数据，大小n*m
    dataSetSize = dataSet.shape[0]
    #求x与各数据坐标的差，形成n*m的差值数组
    diffMat = np.tile(inX, (dataSetSize,1)) - dataSet
    sqDiffMat = diffMat ** 2 
    #np.sum(axis = 0/1)；python自带的sum(sequence,[start,])求和再加start
    sqDistances = sqDiffMat.sum(axis = 1)
    distances = sqDistances ** 0.5
    #np.argsort(x),返回排序升序序号。http://blog.csdn.net/maoersong/article/details/21875705
    sortedDistIndicies = distances.argsort()
    classCount = {}
    for i in range(k):
        voteLabel = labels[sortedDistIndicies[i]]
        classCount[voteLabel] = classCount.get(voteLabel, 0) +1
    #http://www.cnblogs.com/zle1992/p/6271105.html
    #http://blog.csdn.net/dongtingzhizi/article/details/12068205
    #key = operator.itemgetter(1) 提取classCount第2个域作为排序
    #python3 中 用dict.items()没有iteritems()方法了
    #key = operator.itemgetter(1)
    sortedClassCount = sorted(classCount.items(), key = lambda x:x[1], reverse = True)
    #sortedClassCount[0]为行最小第一项，第二个[0]提取第一项的列class
    return sortedClassCount[0][0]
#定义文本txt数据转换numpy格式
def file2matrix(filename):
    with open(filename,'r') as fr:
        #读取全文readlines()
        arrayLines = fr.readlines()
        numberOfLines = len(arrayLines)
        #生成与文件相同大小的zeros
        returnMat = np.zeros(((numberOfLines),3))
        classLabelVector = []
        index = 0
        for line in arrayLines:
            #去空格
            line = line.strip()
            #去line中3项之间的'\t'
            listFromLine = line.split('\t')
            #是否可以for index,line in enumerate(arrayLines):提取index而不单用index
            returnMat[index,:] = listFromLine[0:3]
            #生成classLabel
            classLabelVector.append(int(listFromLine[-1]))
            index += 1
        return returnMat,classLabelVector
datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')
datingDataMat
array([[  4.09200000e+04,   8.32697600e+00,   9.53952000e-01],
       [  1.44880000e+04,   7.15346900e+00,   1.67390400e+00],
       [  2.60520000e+04,   1.44187100e+00,   8.05124000e-01],
       ..., 
       [  2.65750000e+04,   1.06501020e+01,   8.66627000e-01],
       [  4.81110000e+04,   9.13452800e+00,   7.28045000e-01],
       [  4.37570000e+04,   7.88260100e+00,   1.33244600e+00]])

datingLabels[0:20]
[3, 2, 1, 1, 1, 1, 3, 3, 1, 3, 1, 1, 2, 1, 1, 1, 1, 1, 2, 3]

#归一化特征值,newValue = (oldValue - min)/(max - min)
def autoNorm(dataSet):
    minVals = dataSet.min(0)
    maxVals = dataSet.max(0)
    ranges = maxVals - minVals
    #np.shape(a)a是列表或是array都可以提取，也可以array.shape()，但list没有shape属性
    normDataSet = np.zeros(np.shape(dataSet))
    m = dataSet.shape[0]
    #全部减min
    normDataSet = dataSet - np.tile(minVals, (m,1))
    #除以ranges，得到归一化数据
    normDataSet = normDataSet/np.tile(ranges, (m,1))
    return normDataSet, ranges, minVals
normMat, ranges, minVals = autoNorm(datingDataMat)
normMat
array([[ 0.44832535,  0.39805139,  0.56233353],
       [ 0.15873259,  0.34195467,  0.98724416],
       [ 0.28542943,  0.06892523,  0.47449629],
       ..., 
       [ 0.29115949,  0.50910294,  0.51079493],
       [ 0.52711097,  0.43665451,  0.4290048 ],
       [ 0.47940793,  0.3768091 ,  0.78571804]])

def datingClassTest(k):
    #输入待分类数据占训练数据的比例
    hoRatio = 0.1
    #'datingTestSet2.txt'可以运行，'datingTestSet.txt'不能运行
    datingDataMat,datingLabels = file2matrix('datingTestSet2.txt')
    normMat, ranges, minVals = autoNorm(datingDataMat)
    m = normMat.shape[0]
    #分割测试数据与训练数据的位置
    numTestVecs = int(m*hoRatio)
    errorCount = 0.0
    for i in range(numTestVecs):
        #输入数据前numTestVecs个，训练数据dataSet从numTestVecs到m，即剩下的数据集
        classifierResult = classify0(normMat[i,:],normMat[numTestVecs:m,:],\
                                    datingLabels[numTestVecs:m], k)
        print("the classifier came back with: {0},the real answer in :{1}\
        ".format(classifierResult, datingLabels[i]))
        if(classifierResult != datingLabels[i]):
            errorCount += 1.0
    print("the total error rate is: {0}".format(errorCount/float(numTestVecs)))
datingClassTest(3)
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :3        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 3,the real answer in :3        
the classifier came back with: 2,the real answer in :2        
the classifier came back with: 1,the real answer in :1        
the classifier came back with: 3,the real answer in :1        
the total error rate is: 0.05
eclipSYcn
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
机器学习实战--KNN 算法 笔记
木琦
11-07 3501
print "the total error rate is: %f"%(errorCount/float(numTestVecs)) /* 错误率 /*最后得出结果 5%的错误率 机器学习实战书上算的2.4% /* 在不改变其他值的前提下 改变hoRatio的值 [0.05,0.1,0.2,0.3,0.4,0.5，0.9]对应的错误率[2%,5%,8%,8%,7%,6%,7%] 9:在测试结束，我们力求得出一个对于未来的约会预测函数，在我们输入飞行里程数，玩视频游戏的百分比和冰激凌公升数，我们希望
机器学习实战笔记--kNN
sinat_26230689的博客
05-10 707
本文为《机器学习实战》学习笔记 1. 相关数据类型&函数介绍SciPy 基于Python生态系统提供了数学运算、科学和工程的开源软件，主要包括基本N维数组包NumPy，科学计算基本库SciPy library， 用于2D绘图的Matplotlib，交互式控制台IPython，用于符号数学Sympy，用于数据结构和分析的pandas。1.1 NumPyNumPy是python科学计算的基础包。包括强
机器学习实战-kNN
RounrounZhang的博客
01-23 217
读书笔记-机器学习实战-kNN kNN思想：   已知训练样本集和其对应的标签，对于没有标签的新数据，将新数据的每个特征和训练样本集中数据对应的特征进行比较，提取样本集中最相似的k个数据的标签，选择k个最相似数据中出现最多次数的标签作为新数据的标签。 1.从文本文件中解析数据 对未知类别属性的数据集中的每个点依次执行以下操作: (1)计算已知类别数据集中的点与当前点之间的距离
机器学习实战--KNN
进击的码力的博客
10-21 157
按照机器学习实战这本书学习的，本人小白。 之前没有接触过python，所以会穿插自己对于python语法的学习。 所用ide为Pycharm社区版，win，python3.6 如书：首先在KNN.py中增加如下代码 from numpy import* import operator def creatDataSet(): group =array([[1.0,1.1],[1.0,1.0...
机器学习实战4--KNN
FUXI_Willard
04-17 412
本系列博客基于温州大学黄海广博士的机器学习课程的笔记，小伙伴们想更详细学习黄博士课程请移步到黄博士的Github、或者机器学习初学者公众号，现在在中国慕课也是可以学习的，内容包括机器学习、深度学习及Python编程，matplotlib、numpy、pandas、sklearn等，资料很详细，要系统学习请移步哦！笔者的博客只是笔记，内容不会十分详细，甚至会有些少错误！ KNN理论内容：KNN理论基础. 1.概述 kkk近邻法是基本且简单的分类与回归方法。kkk近邻法的基本做法是：对给定的训练实例点和
机器学习实战-kNN分类手写数字笔记
eclipsycn的博客
08-22 387
import numpy as np import pandas as pddef classify0(inX, dataSet, labels, k): #inX代表输入的数据坐标，DataSet:n个m维数据，大小n*m dataSetSize = dataSet.shape[0] #求x与各数据坐标的差，形成n*m的差值数组 diffMat = np.tile(
机器学习实战<学习笔记>--KNN
TomDog的专栏
03-28 533
简单的KNN算法#添加编码方式 # -*- coding: utf-8 -*- from numpy import * import operator# 准备数据 def createDataSet(): group = array([[1.0, 1.1], [1.0, 1.0], [0, 0], [0, 0.1]]) labels = ['A', 'A', 'B', 'B']
机器学习实战笔记--kNN.py模块导入
sunpingblue的博客
05-10 3919
我使用的是Python自带的IDLE写的代码在kNN.py增加如下代码from numpy import *import operatordef createDataSet(): group = array([1.0,1.1],[1.0,1.0],[0,0],[0,0.1]) labels = ['A','A','B','B'] return group,labels保存这个py文件之后&gt;&...
【读书笔记】机器学习实战-kNN(2)
iboxty的专栏
04-11 1249
接着上一篇（这里），把这一章学习完。kNN算法其实并没有其他机器学习算法的“训练算法”的过程，他的执行过程更像是一种统计的过程。他是一种基于实例的学习算法，需要存储所需要的数据集。如果数据集很大将会需要很大的存储空间，时间开销也肯呢个会非常大。这一张的第二个例子是识别手写数字。书已经给出了trainingDigits和testDigits两个文件夹，里面是0~9每个数字的二进制”图像”的txt，实际
《机器学习实战》kNN学习笔记（一）
01-20
《机器学习实战》中的k-近邻(kNN)算法是一种基于实例的学习方法，它通过比较新样本与已知类别样本的距离来决定新样本的类别。本文主要探讨kNN算法的基本概念、优缺点以及一般流程，并提供一个简单的Python实现示例。...
python unicode 及解码编码方式简介
热门推荐
eclipsycn的博客
12-19 1万+
python及编码原理测试基于utf—8环境。#coding:utf-8 unicode为通用编码。coding:utf-8的作用是声明python解释器及str的编码方式，并不改变其他sys.getdefaultencoding()的默认编码方式。unicode 经过encode变为 strstr 经过decode变为 unicodeu’你好’ 代表 unicode 形式的你好 utf
人脸识别face recognition
eclipsycn的博客
10-29 8105
0 简介 人脸识别大致分为如下三个部分： 人脸检测 face detection 人脸对齐 face alignment 人脸识别 face recognition，包括： face verification：两张图片相似程度。 face identification： 在图片库中检索与当前图片相似度最高的图片。 1 人脸检测 face detection通常认为是在图片中找到人
module 'pandas' has no attribute 'computation'
eclipsycn的博客
08-18 7674
若是Anaconda安装的包，更新dask即可 conda update dask
python list赋值不改变
eclipsycn的博客
01-04 7044
list赋值在修改facenet程序时，发现用摄像头中出现两个人脸的时候，每次只能框一个人脸，而5个特征点则两个人都能显示。经过一番筛查，是在识别人脸的程序中，list初始化 bb = np.zeros(4,dtype=np.int32)这句话应该放在每次循环的里面。放在外面导致bb变化以后，tbb.append(bb)发现tbb[0] = tbb[1]。而以前都是直接切片获取 cropface =
python 装饰器@及带参数的情况
eclipsycn的博客
02-02 4391
一般装饰器：def use_logging(func): def wrapper(*args, **kwargs): if level == 'warn': logging.warn('%s is running' % func.__name__) return func(*args) return wrapper@use_lo
利用python读取csv中url并下载
eclipsycn的博客
02-27 3902
import urllib2 import pandas as pd import os 导入pandas库读取csv文件 data = pd.read_csv('train_attrs(2).csv') data.head() id img_url xml_url ...
自动下载cs231n课件
eclipsycn的博客
08-13 2322
cs231n 2017年课件地址 http://cs231n.stanford.edu/slides/2017/创建好一个文件夹，例如在桌面上创建cs231 lecture，路径为 C:\Users\Administrator\Desktop\cs231n lecture 打开cmd， cd C:\Users\Administrator\Desktop\cs231n lecture jupyt
Resnet-Tensorflow 在Mask-Rcnn 中的结构
eclipsycn的博客
12-20 2054
阅读Mask-Rcnn代码时，观察resnet结构。 其中重要的有identity_block和conv_block。Identity Block:1*1，3*3，1*1的三层stride=(1,1)卷积，最后直接输出x+input作为输出。def identity_block(input_tensor, kernel_size, filters, stage, block,
机器学习实战：KNN算法应用与参数优化
本篇文档主要介绍了机器学习中的KNN分类实验，结合Python库sklearn进行实战操作。实验的目标在于深入理解KNN分类的基本原理，并通过实践掌握如何使用该算法以及调整其参数。首先，实验要求参与者在计算机上实现KNN...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值