import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.metrics import *
import random
import operator

def knn(x_test,x_data,y_data,k= 5):
    x_data_size = x_data.shape[0]
    new_x_test = np.tile(x_test,(x_data_size,1))
    diffMat = new_x_test - x_data
    sqdiffmat = diffMat**2
    distance = sqdiffmat .sum(axis=1)**0.5
    sortDistance = distance.argsort()
    class_num={}
    for i in range(k):
        notelabel=y_data[sortDistance[i]]
        class_num[notelabel]=class_num.get( notelabel,0)+1

    sorted_class_num = sorted(class_num.items(),key=operator.itemgetter(1),reverse=1)
    knnclass = sorted_class_num[0][0]
    return knnclass
    
def main():
    iris_data= datasets.load_iris()
    data = iris_data.data

    #*************混ぜるデータ・打乱数据*************
    index = [i for i in range(data.shape[0])]
    random.shuffle(index)
    data = data[index]
    iris_data.target = iris_data.target[index]

    train_data_num = int(data.shape[0]*0.8)
    test_data_num = data.shape[0]-int(data.shape[0]*0.8)

    x_train_data = data[:train_data_num]
    y_train_data =iris_data.target[:train_data_num]
    x_test_data =data[train_data_num:]
    y_test_data =iris_data.target[train_data_num:]

    print("train_data count = ",train_data_num)
    print("test_data count = ",test_data_num)

    predictions =[]

    for i in range(test_data_num):
        predictions.append(knn(x_test_data[i],x_train_data,y_train_data))

    #print (predictions)

    print (classification_report(y_test_data,predictions))
    print (confusion_matrix(y_test_data,predictions))


main()

鸢尾花_KNN分类_数据