K Nearest Neibours

最新推荐文章于 2024-09-19 12:23:48 发布

Yukei7

最新推荐文章于 2024-09-19 12:23:48 发布

阅读量266

点赞数

CC 4.0 BY-SA版权

分类专栏：学习笔记文章标签：机器学习

本文链接：https://blog.youkuaiyun.com/qq373271567/article/details/84900970

学习笔记专栏收录该内容

1 篇文章

订阅专栏

本文深入讲解了K近邻(KNN)算法的实现原理，通过Python代码演示如何创建数据集、计算距离并进行分类。KNN是一种基本的监督学习算法，适用于分类和回归任务。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

from numpy import *
import operator

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

def classify0(inX, dataSet, labels, k):
    dataSetSize = dataSet.shape[0] # shape[0]:get the number of rows of the dataset matrix (i.e.size of the dataset)
    diffMat = tile(inX, (dataSetSize,1)) - dataSet # inX:unlabelled data, get the difference of inX and dataSet(each element in dataSet)
    sqDiffMat = diffMat ** 2 # return square of each element in diffMat
    sqDistances = sqDiffMat.sum(axis=1) # axis=n :remove dimension n then return sum?
    distances = sqDistances ** 0.5 # Euclid distance
    sortedDisIndicies = distances.argsort() #return index of the sorted number(from small to large)
    classCount={}
    for i in range(k):
        voteIlabel = labels[sortedDisIndicies[i]] # get the nearest k number lables
        classCount[voteIlabel] = classCount.get(voteIlabel,0) + 1 #find the correspond label(if there is not, append '0' to the dict)
    sortedClassCount = sorted(classCount.items(),
        key=operator.itemgetter(1), reverse=True)
    return sortedClassCount[0][0] # the label with most votes