优点:精度高,对异常值不敏感,无数据输入假定
缺点:计算复杂度高,空间复杂度高
适用数据范围:数值型和标称型
 
一般流程:
    (1). 收集数据(网络抓取)
    (2).处理数据,将数据处理成结构化的数据格式。
    (3).分析数据
    (4).测试算法(主要是计算模型的出错率)
    (5).使用算法,
 
K-近邻算法采用测量不同特征值之间的距离的方法进行分类
 
工作原理是:存在一个训练样本集,且样本集中每个数据都存在标签(与分类的对应关系)。
当输入没有标签的新数据后,将新数据的每个特征与样本集中的数据对应的特征进行比较,
然后算法提取训练样本集中前k个最相似的数据的分类标签,且k不大于20 。
选择最相似数据中出现次数最多的分类,作为新数据的分类。
 
 
数据归一化的作用,只用在特征数据相差较大且同等重要的条件下
 
aaarticlea/png;base64,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" alt="" name="en-media:image/png:2bdbd1d41ecee133d740c53250e97a8b:none:none" />
 
上面方程中数字差值最大的属性对计算结果的影响最大,仅仅是因为飞行常客里程数远大于其他特征值。然而我们认为这三种特征同样重要,因此作为三个等权重的特征 
 
直接上代码:
from numpy import *
import matplotlib.pyplot as plot
import operator
from os import listdir def classify0(inX, dataSet, labels, k):
dataSetSize = dataSet.shape[0]
# 距离计算公式
diffMat = tile(inX, (dataSetSize,1)) - dataSet
sqDiffMat = diffMat**2
sqDistances = sqDiffMat.sum(axis=1)
distances = sqDistances**0.5 # 距离从大到小排序,返回距离的序号
sortedDistIndicies = distances.argsort()
# 声明一个空的字典,用于存放标签
classCount={}
for i in range(k):
# sortedDistIndicies[0]返回的是距离最小的数据样本的序号
# labels[sortedDistIndicies[0]]距离最小的数据样本的标签
voteIlabel = labels[sortedDistIndicies[i]]
classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1
# 给该字典排序,sortedClassCount[0][0]是K中支持的标签数最大的
sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)
print(sortedClassCount[0][0])
return sortedClassCount[0][0] # 创建数据
def createDataSet():
group = array([[1.0,1.1],[1.0,1.0],[0,0],[0,0.1]])
labels = ['A','A','B','B']
return group, labels # 画图
def draw(xs,ys):
fig = plot.figure()
# 将画布分割成1行1列,图像画在从左到右从上到下的第1块
# 设置画布的大小与图像的位置
ax = fig.add_subplot(221)
# ax.scatter(xs, ys)的两个参数分别是所有点的x坐标,所有点的y坐标
ax.scatter(xs,ys)
plot.show() def firstTest():
test1 = (1.0, 1.2)
test2 = (0.0, 0.4)
dataset, labels = createDataSet()
conclusion1 = classify0(test1, dataset, labels, 3)
conclusion2 = classify0(test2, dataset, labels, 3)
print(str(test1) + "分类后的结果是属于" + conclusion1 + "类")
print(str(test2) + "分类后的结果是属于" + conclusion2 + "类")
# 将32*32的矩阵读为1*1024
def img2vector(filename):
returnVect = zeros((1,1024))
fr = open(filename)
for i in range(32):
lineStr = fr.readline()
for j in range(32):
returnVect[0,32*i+j] = int(lineStr[j])
return returnVect def handwritingClassTest():
hwLabels = []
# 获得训练样本数据集
trainingFileList = listdir('digits/trainingDigits')
# 样本数的个数
m = len(trainingFileList)
# 返回m行1024列的矩阵数据
trainingMat = zeros((m, 1024))
# 文件名下划线_左边的数字是标签
for i in range(m):
fileNameStr = trainingFileList[i]
fileStr = fileNameStr.split(".")[0]
# 分类标签
classNumStr = int(fileStr.split('_')[0])
hwLabels.append(classNumStr)
trainingMat[i, :] = img2vector('digits/trainingDigits/%s' % fileNameStr)
testFileList = listdir('digits/testDigits')
errorCount = 0.0
mTest = len(testFileList)
for i in range(mTest):
fileNameStr = testFileList[i]
fileStr = fileNameStr.split('.')[0] # take off .txt
classNumStr = int(fileStr.split('_')[0])
vectorUnderTest = img2vector('digits/testDigits/%s' % fileNameStr)
classifierResult = classify0(vectorUnderTest, trainingMat, hwLabels, 3)
print("the classifier came back with: %d, the real answer is: %d" % (classifierResult, classNumStr))
if (classifierResult != classNumStr): errorCount += 1.0
print("\nthe total number of errors is: %d" % errorCount)
print("\nthe total error rate is: %f" % (errorCount / float(mTest))) # 主函数调用模块函数
if __name__ == "__main__":
# group,label = createDataSet()
# # group[:, 0] 所有行的第0列
# draw(group[:, 0], group[:, 1])
# # print(group[:, 0])
# firstTest()
handwritingClassTest() 训练数据集合测试集的数据:https://gitee.com/lcl1993213/plist

KNN--用于手写数字识别的更多相关文章

  1. KNN实现手写数字识别

    KNN实现手写数字识别 博客上显示这个没有Jupyter的好看,想看Jupyter Notebook的请戳KNN实现手写数字识别.ipynb 1 - 导入模块 import numpy as np i ...

  2. Softmax用于手写数字识别(Tensorflow实现)-个人理解

    softmax函数的作用   对于分类方面,softmax函数的作用是从样本值计算得到该样本属于各个类别的概率大小.例如手写数字识别,softmax模型从给定的手写体图片像素值得出这张图片为数字0~9 ...

  3. 机器学习(二)-kNN手写数字识别

    一.kNN算法是机器学习的入门算法,其中不涉及训练,主要思想是计算待测点和参照点的距离,选取距离较近的参照点的类别作为待测点的的类别. 1,距离可以是欧式距离,夹角余弦距离等等. 2,k值不能选择太大 ...

  4. 一看就懂的K近邻算法(KNN),K-D树,并实现手写数字识别!

    1. 什么是KNN 1.1 KNN的通俗解释 何谓K近邻算法,即K-Nearest Neighbor algorithm,简称KNN算法,单从名字来猜想,可以简单粗暴的认为是:K个最近的邻居,当K=1 ...

  5. kaggle 实战 (1): PCA + KNN 手写数字识别

    文章目录 加载package read data PCA 降维探索 选择50维度, 拆分数据为训练集,测试机 KNN PCA降维和K值筛选 分析k & 维度 vs 精度 预测 生成提交文件 本 ...

  6. Kaggle竞赛丨入门手写数字识别之KNN、CNN、降维

    引言 这段时间来,看了西瓜书.蓝皮书,各种机器学习算法都有所了解,但在实践方面却缺乏相应的锻炼.于是我决定通过Kaggle这个平台来提升一下自己的应用能力,培养自己的数据分析能力. 我个人的计划是先从 ...

  7. 基于OpenCV的KNN算法实现手写数字识别

    基于OpenCV的KNN算法实现手写数字识别 一.数据预处理 # 导入所需模块 import cv2 import numpy as np import matplotlib.pyplot as pl ...

  8. K近邻实战手写数字识别

    1.导包 import numpy as np import operator from os import listdir from sklearn.neighbors import KNeighb ...

  9. C#中调用Matlab人工神经网络算法实现手写数字识别

    手写数字识别实现 设计技术参数:通过由数字构成的图像,自动实现几个不同数字的识别,设计识别方法,有较高的识别率 关键字:二值化  投影  矩阵  目标定位  Matlab 手写数字图像识别简介: 手写 ...

  10. 利用神经网络算法的C#手写数字识别

    欢迎大家前往云+社区,获取更多腾讯海量技术实践干货哦~ 下载Demo - 2.77 MB (原始地址):handwritten_character_recognition.zip 下载源码 - 70. ...

随机推荐

  1. VS2017生成解决方案报错,提示对路径的访问被拒绝

    目前我用的vs2017的版本是15.3.5.生成解决方案有时会提示如下: 开始以为是权限的问题,找到相应的目录设置everyone权限,再次生成还是不行.重启VS试了下,还是不行. 最后无奈重启下电脑 ...

  2. LeetCode 191. Number of 1 bits (位1的数量)

    Write a function that takes an unsigned integer and returns the number of ’1' bits it has (also know ...

  3. sklearn中各算法类的fit,fit_transform和transform函数

    在使用PCA和NFC中有三个函数fit,fit_transform,transform区分不清各自的功能.通过测试,勉强了解各自的不同,在这里做一些笔记. 1.fit_transform是fit和tr ...

  4. undefined 与void 0

    参考:https://segmentfault.com/a/1190000000474941 Javascript中void是一个操作符,该操作符指定要计算一个表达式但是不返回值.void 操作符用法 ...

  5. JAVAscript学习笔记 js条件语句 第三节 (原创) 参考js使用表 (2017-09-14 15:55)

    <!DOCTYPE html> <html lang="en"> <head> <meta charset="UTF-8&quo ...

  6. [java基础] 遇到的一个关于返回值泛型的问题

    在写代码的时候这样写: import java.util.ArrayList; import java.util.List; public class TestConversion { public ...

  7. C#数据类型和SQL数据类型对照

    C#操作SQL Float类型,数据会多很多小数,原来是C#的float和sql的float类型不一致        /// <summary>        /// 数据库中与C#中的数 ...

  8. 为什么选择使用Sass而不是Less?

    这篇文章主要解答以下几个问题,供前端开发者的新手参考. 1.什么是Sass和Less? 2.为什么要使用CSS预处理器? 3.Sass和Less的比较 4.为什么选择使用Sass而不是Less? 什么 ...

  9. Nodejs mongodb 管理组件adminmongodb

    强大的 nodejs的mongodb管理工具,强大到即下即用: 安装需求: 1.git命令获取组件包,git clone https://github.com/mrvautin/adminMongo. ...

  10. Python之random

    random 伪随机数生成模块.如果不提供seed,默认使用系统时间. 使用相同seed,可获得相同的随机数序列,常用于测试. >>> from random import * &g ...