knn 鸢尾花案列

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn.preprocessing import StandardScaler

def get_data(fire_path):    # 获取数据,存为df
    file = pd.read_csv('Iris.csv') 
    df = pd.DataFrame(file)
    return df
def check_data(df):  # 查看数据信息
    print(df.head())
    print(df.tail())
    print(df.shape)
    print(df.describe)
    print(df.info)
def prepro_data(df):  # 处理数据，转为为字典
    arr2 = np.array(df['Species'])
    df.drop('Species',axis = 1,inplace = True) # 去掉标签列
    df.drop('Id',axis = 1,inplace = True)     # 去掉id列
    arr = np.array(df)     # 获取数据转化成数组
    iris = {'data':arr,'target':arr2}
    return iris
def eng_data(iris):   # 特征工程
    X = iris['data']  # 获取数据
    Y = iris['target'].reshape(-1,1)  # 标签转换成数字
    #print(x.shape,y.shape)  # 数据和标签的格式
    x_train, x_test, y_train, y_test = train_test_split(X, Y , test_size=0.3, random_state=22)  # 分割训练集和测试集，比例3:7
    TF = StandardScaler()   # 实例化
    x_train = TF.fit_transform(x_train)  #特征值
    x_test = TF.fit_transform(x_test)   # 目标值
    return x_train,x_test,y_train, y_test
def training(x_train,y_train,n_neighbors):  # 训练模型
    estimator = KNeighborsClassifier(n_neighbors)
    estimator.fit(x_train, y_train)
    return estimator
def asse(estimator,x_test,y_test):  # 预测评估
    y_predict = estimator.predict(x_test)
    print("预测值真实值的对比：\n", y_predict == y_test)
    score = estimator.score(x_test, y_test)  # x_代表测试集数据，y_test代表测试集标签。
    print(score)
    
fire_path = 'Iris.csv'
n_neighbors = 3   # 设置邻近

df = get_data(fire_path)  # 1，从scv文件获取数据，存为df
#print(df)

iris = prepro_data(df)    # 2，数据预处理，去掉不用的列，将数据转为字典
#print(iris)

x_train,x_test,y_train, y_test = eng_data(iris)   # 3，特征工程，传入字典，将数据进行分割，并进行归一化与标准化处理,返回处理好的数据
#print(x_train,x_test,y_train, y_test)

estimator = training(x_train,y_train,n_neighbors)  # 4，传入数据，进行训练，这里可以设置n_neighbors的值进而改变准确率
#print(estimator)

asse(estimator,x_test,y_test)                       #5， 传入训练的模型，测试数据，进行评估与预测

准确率