基于51单片机音乐盒设计( proteus仿真+程序+原理图+PCB+报告+讲解视频）

文章目录

• 数据准备阶段
• KNN预测的过程
• • 1.计算新样本与已知样本点的距离
  • 2.按照举例排序
  • 3.确定k值
  • 4.距离最近的k个点投票
• scikit-learn中的KNN算法

数据准备阶段

import matplotlib.pyplot as plt
import numpy as np

# 样本特征
data_X = [
    [0.5, 2],
    [1.8, 3],
    [3.9, 1],
    [4.7, 4],
    [6.2, 6],
    [7.5, 5],
    [8.3, 3.5],
    [9.1, 7],
    [9.8, 4.5]
]

# 样本标记
data_y = [0, 0, 0, 1, 1, 1, 1, 1, 1]

X_train = np.array(data_X)
y_train = np.array(data_y)

X_train

y_train

选出样本标记为0的样本特征

y_train == 0

X_train[y_train==0]

X_train[y_train==0, 0]

X_train[y_train==0, 1]

X_train[y_train==1, 0].shape

X_train[y_train==1, 1].shape

plt.scatter(X_train[y_train==0, 0], X_train[y_train==0, 1], color='red', marker='x')
plt.scatter(X_train[y_train==1, 0], X_train[y_train==1, 1], color='black', marker='o')
plt.show()

增加新的样本点

data_new = np.array([4, 5])

plt.scatter(X_train[y_train==0, 0], X_train[y_train==0, 1], color='red', marker='x')
plt.scatter(X_train[y_train==1, 0], X_train[y_train==1, 1],color='black', marker='o')
plt.scatter(data_new[0], data_new[1], color='b', marker='^')
plt.show()

KNN预测的过程

1.计算新样本与已知样本点的距离

for data in X_train:
    print(np.sqrt(np.sum((data - data_new) ** 2)))

distances = [np.sqrt(np.sum((data - data_new) ** 2)) for data in X_train]
distances

2.按照举例排序

np.sort(distances)

sort_index = np.argsort(distances)
sort_index

3.确定k值

k = 5

4.距离最近的k个点投票

first_k = [y_train[i] for i in sort_index[:k]]
first_k

from collections import Counter

Counter(first_k)

Counter(first_k).most_common()

Counter(first_k).most_common(1)

predict_y = Counter(first_k).most_common(1)[0][0]
predict_y

得到结果为1，KNN判断新加入的点data_y的标记应该为1，从图中也可以看到，新加入的点更靠近标记为1的点群。

scikit-learn中的KNN算法

 from sklearn.neighbors import KNeighborsClassifier

kNN_classifier = KNeighborsClassifier(n_neighbors=5)

kNN_classifier.fit(X_train, y_train)

data_new.reshape(1, -1)

predict_y = kNN_classifier.predict(data_new.reshape(1, -1))
predict_y

与手写KNN得到的结果相同，皆判断为1。