ch02_knn

第二章.knn

import numpy as np
import operator

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

# 下面代码使用knn进行一次简单的分类
def classify0(input_data, data_set, labels, k):
    # 1、计算距离
    # 2、找到最近的k个数据
    # 3、统计他们的类别
    # 4、返回类别最多的那个
    # 1、
    input_data = (input_data - data_set) ** 2
    distances = np.sqrt(np.sum(input_data, axis=1))

    # 2+3
    sorted_data_index = distances.argsort()
    class_count = {}
    for i in range(k):
        label = labels[sorted_data_index[i]]
        class_count[label] = class_count.get(label, 0) + 1
    
    sorted_class_count = sorted(class_count.items(), key = lambda x: x[1], reverse=True)
    return sorted_class_count[0][0]

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

'B'

例子1：使用knn进行约会网站的配对

00.使用pandas处理数据

import pandas as pd
df = pd.read_csv("dataset/datingTestSet2.txt", sep='\t', names=['games', "miles", "ice_cream", "category"])
df

	games	miles	ice_cream	category
0	40920	8.326976	0.953952	3
1	14488	7.153469	1.673904	2
2	26052	1.441871	0.805124	1
3	75136	13.147394	0.428964	1
4	38344	1.669788	0.134296	1
...	...	...	...	...
995	11145	3.410627	0.631838	2
996	68846	9.974715	0.669787	1
997	26575	10.650102	0.866627	3
998	48111	9.134528	0.728045	3
999	43757	7.882601	1.332446	3

1000 rows × 4 columns

01.文件转变量

# 将文件转换为numpy数组
def file2metrix(filename):
    df = pd.read_csv("dataset/datingTestSet2.txt", sep='\t', names=['games', "miles", "ice_cream", "category"])
    features = np.array([list(df.games.values), list(df.miles.values), list(df.ice_cream.values)]).T
    category = np.array(list(df.category.values))

    return features, category

features, category = file2metrix("dataset/datingTestSet2.txt")
category[:20]

array([3, 2, 1, 1, 1, 1, 3, 3, 1, 3, 1, 1, 2, 1, 1, 1, 1, 1, 2, 3],
      dtype=int64)

02.分析数据

# 图像显示中文字体
import matplotlib.pyplot as plt
plt.rcParams["font.sans-serif"]=["SimHei"] #设置字体
plt.rcParams["axes.unicode_minus"]=False #该语句解决图像中的“-”负号的乱码问题

plt.scatter(features[:, 1], features[:, 2], c=category)
plt.xlabel("游戏时间")
plt.ylabel("冰淇淋公升数")
plt.show()

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-uoVHGR5N-1665481867492)(output_12_0.png)]

category_1_mask = category == 1
plt.plot(features[:, 0][category_1_mask], features[:, 1][category_1_mask], 'bo', label="不喜欢")
category_2_mask = category == 2
plt.plot(features[:, 0][category_2_mask], features[:, 1][category_2_mask], 'go', label="魅力一般")
category_3_mask = category == 3
plt.plot(features[:, 0][category_3_mask], features[:, 1][category_3_mask], 'ro', label="极具魅力")
plt.legend(loc="best")
plt.xlabel("里程数")
plt.ylabel("游戏时间")
plt.show()

[外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传(img-I57OXiFG-1665481867493)(output_13_0.png)]

03.归一化数据

def autoNorm(dataset):
    # 将数据归一化
    max_value = np.max(dataset, axis=0)
    min_value = np.min(dataset, axis=0)

    ranges = max_value - min_value

    dataset_normilized = (dataset - min_value) / ranges

    return dataset_normilized, ranges, min_value

dataset_normilized, ranges, min_value = autoNorm(features)
min_value

array([0.      , 0.      , 0.001156])

04.测试算法

# 将一部分数据作为测试数据
def dating_class_test():
    test_ratio = 0.1

    # 1、获取数据
    features, labels = file2metrix("dataset/datingTestSet2.txt")

    # 2、对数据进行归一化操作
    features_normilized = autoNorm(features)

    # 3、取一部分作为测试数据
    data_size = features.shape[0]
    random_index = np.arange(data_size)
    np.random.shuffle(random_index)

    test_data_size = int(data_size * test_ratio)
    features_test = features[random_index[:test_data_size]]
    features_train = features[random_index[test_data_size:]]
    labels_train = labels[random_index[test_data_size:]]
    labels_test = labels[random_index[:test_data_size]]

    error_count = 0

    for i in range(test_data_size):
        label_pred = classify0(features_test[i], features_train, labels_train, 3)
        if label_pred != labels_test[i]:
            error_count += 1

    error_percent = error_count / float(test_data_size)
    print("错误率是：{0}%".format(error_percent)) 

dating_class_test()

错误率是：0.22%

05.构建可用系统

numpy.ndarray

def classify_person():
    result_labels = ['不喜欢', '小魅力', '大魅力']
    dating_features, dating_labels = file2metrix("dataset/datingTestSet2.txt")
    dating_features_normilized, ranges, min_value = autoNorm(dating_features)

    input_game = input("输入他每周花费在游戏上的时间")
    input_miles = input("输入他每周飞行的里程数")
    input_ice_cream = input("输入他每周吃的冰淇淋公升数")
    input_feature = (np.array([float(input_game), float(input_miles), float(input_ice_cream)]) - min_value) / ranges 

    classified_result = classify0(input_feature, dating_features_normilized, dating_labels, 3)
    print("这个人的类型属于：", result_labels[classified_result - 1]) 

classify_person()

这个人的类型属于： 不喜欢