原理：计算当前点（无label，一般为测试集）和其他每个点（有label，一般为训练集）的距离并升序排序，选取k个最小距离的点，根据这k个点对应的类别进行投票，票数最多的类别的即为该点所对应的类别。
代码实现（数据集采用的是iris）：
1
import
numpy as np
2
from sklearn.datasets import load_iris
3from sklearn.model_selection import train_test_split
4from sklearn import neighbors
5from sklearn.metrics import accuracy_score
6 7def get_iris():
8 iris_data = load_iris()
9 X_train, X_test, y_train, y_test = train_test_split(iris_data.data, iris_data.target, test_size=0.4, random_state=0)
10return X_train, X_test, y_train, y_test
1112def knn_classify(self_point, dataset, labels, k):
13 distance = [np.sqrt(sum((self_point - d)**2)) for d in dataset]
14 train_data = zip(distance, labels)
15 train_data = sorted(train_data, key=lambda x: x[0])[:k]
16 self_label = {}
17for i in train_data:
18 i = str(i[1])
19 self_label[i] = self_label.setdefault(i, 0) + 1
20 self_label = sorted(self_label, key=self_label.get, reverse=True)
21return self_label[0]
222324 X_train, X_test, y_train, y_test = get_iris()
25 size = len(y_test)
26 count = 0
27for t in range(len(X_test)):
28 y_pre = knn_classify(X_test[t], X_train, y_train, 5)
29if y_pre == str(y_test[t]):
30 count += 1
31print(‘custom的准确率： ‘, count / size)
3233# 使用sklearn内置的KNN34 knn = neighbors.KNeighborsClassifier(n_neighbors=5)
35knn.fit(X_train, y_train)
36 pre = knn.predict(X_test)
37print(‘sklearn的准确率: ‘, accuracy_score(y_test, pre))
对比结果：
custom的准确率： 0.95
sklearn的准确率: 0.95
原文：http://www.cnblogs.com/laresh/p/7586625.html
【说明】：本文章由站长整理发布，文章内容不代表本站观点，如文中有侵权行为，请与本站客服联系(QQ:254677821)！