k-近邻算法

以机器学习实战上面一段代码为例

import numpy as np
import operator

"""创建数据集"""
def createDataSet():
    group = np.array([[1.0, 1.1], [1.0, 1.0], [0, 0], [0, 0.1]])
    labels = ['A', 'A', 'B', 'B']
    return group, labels

 

 

"""k-近邻算法"""
def classify0(X, dataset, labels, k):
    """
    X:用于输入的待分类向量
    dataset:训练样本特征数据集
    labels：训练样本对应的标签数据集
    """
    # 训练数据的样本个数
    dataset_size = dataset.shape[0]
    # 计算x与训练集每个样本的距离(此处选择欧氏距离)
    diffMat = np.tile(X, (dataset_size, 1)) - dataset
    sqDiffMat = diffMat**2
    sqDistances = np.sum(sqDiffMat, axis = 1)
    distances = sqDistances**0.5
    # 获得与样本距离从小到大排列后在愿标签中的的位置索引
    sortedDistIndicies = np.argsort(distances)

    # 获得与X最近的k个训练样本类别出现的次数
    classCount = {}
    for i in range(k):
        # 获得离X第i近的样本标签
        voteIlabel = labels[sortedDistIndicies[i]]
        # 对该标签进行统计累加次数
        classCount[voteIlabel] = classCount.get(voteIlabel, 0) + 1
    # 对字典sortedClassCount按值从大到小进行排序
    sortedClassCount = sorted(classCount.items(), key=lambda x: x[1], reverse = True)
    # 返回X的类别
    return sortedClassCount[0][0]

group, labels = createDataSet()
print(group)
print(labels)

 out:

[[1.  1.1]
 [1.  1. ]
 [0.  0. ]
 [0.  0.1]]
['A', 'A', 'B', 'B']

classify0([0,0], group, labels, 3)

out:

'B'