距离度量+特征预处理

                                                                            距离度量

1. 欧氏距离：两个点在空间中的距离一般都是指欧氏距离

2. 曼哈顿距离：称为“城市街区距离.            曼哈顿城市特点：横平竖直

3. 切比雪夫距离:二个点之间的距离定义是其各坐标数值差绝对值的最大值

4. 闵可夫斯基距离:不是一种新的距离的度量方式

                              是对多个距离度量公式的概括性的表述

                                                                   特征预处理

1.  为什么做归一化和标准化?

特征的单位或者大小相差较大，或者某特征的方差相比其他的特征要大出几个数量级，容易影响（支配）目标结果，使得一些模型（算法）无法学习到其它的特征。

归一化：通过对原始数据进行变换把数据映射到【mi,mx】(默认为[0,1])之间

标准化：通过对原始数据进行标准化，转换为均值为0标准差为1的标准正态分布的数据

加载鸢尾花数据集

​​​from sklearn.datasets import load_iris
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
mydataset = load_iris()
print(mydataset.feature_names)
iris_d= pd.DataFrame(mydataset['data'], columns=mydataset.feature_names)
iris_d['Species'] = mydataset.target
print('\niris_d-->\n', iris_d)
col1 = 'sepal length (cm)'
col2 = 'petal width (cm)'
sns.lmplot(x=col1, y=col2, data=iris_d, hue='Species', fit_reg=False)
plt.xlabel(col1)
plt.ylabel(col2)
plt.title('iris')
plt.show()

数据集划分

​from sklearn.model_selection import train_test_split
from sklearn.datasets import load_iris
mydataset = load_iris()
X_train, X_test, y_train, y_test = train_test_split(mydataset.data, mydataset.target,
test_size=0.3, random_state=22)
print('数据总数量',len(mydataset.data))
print('训练集中的x-特征值',len(X_train))
print('测试集中的x-特征值',len(X_test))
print(y_train)

模型训练

​from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.datasets import load_iris
mydataset = load_iris()
x_train, x_test, y_train, y_test = train_test_split(mydataset.data, mydataset.target, test_size=0.2,
random_state=22)
transfer = StandardScaler()
x_trai = transfer.fit_transformfrom sklearn.preprocessing import StandardScaler(x_train)
x_test = transfer.transform(x_test)
estimator = KNeighborsClassifier(n_neighbors=3)
estimator.fit(x_train, y_train)
myscore = estimator.score(x_test, y_test)
print('myscore-->', myscore)
print('通过模型查看相应类别-->', estimator.classes_)
mydata =[[5.1, 3.5, 1.4, 0.2],
         [4.6, 3.1, 1.5, 0.2]]
mydata = transfer.transform(mydata)
print('mydata-->', mydata)
mypred = estimator.predict(mydata)
print('mypred-->\n', mypred)
mypred = estimator.predict_proba(mydata)
print('mypred-->\n', mypred)

模型评估

​from sklearn.datasets import load_iris
from sklearn.metrics import accuracy_score
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import StandardScaler
mydataset = load_iris()
x_train, x_test, y_train, y_test = train_test_split(mydataset.data, mydataset.target,
test_size=0.2, random_state=22)
transfer = StandardScaler()
x_train = transfer.fit_transform(x_train)
x_test = transfer.transform(x_test)
estimator = KNeighborsClassifier(n_neighbors=3)
estimator.fit(x_train, y_train)
myscore = estimator.score(x_test, y_test)
print('myscore-->', myscore)
y_predict = estimator.predict(x_test)
myresult = accuracy_score(y_test, y_predict)
print('myresult-->', myresult)