kNN分類算法的Python實現

1.k-近鄰算法實現

from numpy import *
import operator


def createDataSet():
    group = array([[1.0, 1.1], [2.0, 2.0], [0, 0], [4.1, 5.1]])
    labels = ['A', 'B', 'C', 'D']
    return group, labels


def classify0(inX, dataSet, labels, k):
    """

    :param inX: 用於分類的輸出向量
    :param dataSet:輸入的樣本集
    :param labels:標簽向量
    :param k:用於選擇最近鄰居的樹目
    :return:
    """
    dataSetsize = dataSet.shape[0]  # 得到數據集的行數
    diffMat = tile(inX, (dataSetsize, 1)) - dataSet  # tile生成和訓練樣本對應的矩陣，並與訓練樣本求差
    sqDiffMat = diffMat ** 2
    sqDistances = sqDiffMat.sum(axis=1)  # 將矩陣的每一行相加
    distances = sqDistances ** 0.5
    sortedDistIndicies = distances.argsort()  # 從小到大排序 返回對應的索引位置
    classCount = {}
    for i in range(k):
        voteIlabel = labels[sortedDistIndicies[i]]  # 找到該樣本的類型
        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1  # 在字典中將該類型加一
    sortedClassCount = sorted(classCount.items(), key=operator.itemgetter(1), reverse=True)  # reverse = True代表降序
    return sortedClassCount[0][0]  # 排序並返回出現最多的那個類型

2.測試

import kNN


group,labels = kNN.createDataSet()
print(kNN.classify0([0,0],group,labels,3))
print(kNN.classify0([1,2],group,labels,3))
print(kNN.classify0([3,3],group,labels,3))
print(kNN.classify0([5,5],group,labels,3))

3.實驗結果

C
A
B
D

實驗環境：Ubuntu18.04+Pycharm+python3.6+numpy