Understanding k-Nearest Neighbour

我們將Red系列標(biāo)記為Class-0（由0表示），將Blue 系列標(biāo)記為Class-1（由1表示）。 我們創(chuàng)建了25個(gè)系列或25個(gè)訓(xùn)練數(shù)據(jù)，并將它們標(biāo)記為0級(jí)或1級(jí).在Matplotlib的幫助下繪制它,紅色系列顯示為紅色三角形，藍(lán)色系列顯示為藍(lán)色方塊.

import numpy as np
import cv2
import matplotlib.pyplot as plt

# Feature set containing (x,y) values of 25 known/training data
trainData = np.random.randint(0,100,(25,2)).astype(np.float32)

# Labels each one either Red or Blue with numbers 0 and 1
responses = np.random.randint(0,2,(25,1)).astype(np.float32)

# Take Red families and plot them
red = trainData[responses.ravel()==0]
plt.scatter(red[:,0],red[:,1],80,"r","^")

# Take Blue families and plot them
blue = trainData[responses.ravel()==1]
plt.scatter(blue[:,0],blue[:,1],80,"b","s")

plt.show()

接下來初始化kNN算法并傳遞trainData和響應(yīng)以訓(xùn)練kNN（它構(gòu)造搜索樹）.然后我們將對(duì)一個(gè)new-comer，并在OpenCV的kNN幫助下將它歸類為一個(gè)系列.KNN之前，我們需要了解一下我們的測試數(shù)據(jù)（new-comer),數(shù)據(jù)應(yīng)該是一個(gè)浮點(diǎn)數(shù)組，其大小為numberoftestdata×numberoffeatures.然后找到new-comer的最近的鄰居并分類.

newcomer = np.random.randint(0,100,(1,2)).astype(np.float32)
plt.scatter(newcomer[:,0],newcomer[:,1],80,"g","o")

knn = cv2.ml.KNearest_create()
knn.train(trainData, cv2.ml.ROW_SAMPLE, responses)
ret, results, neighbours ,dist = knn.findNearest(newcomer, 3)

print( "result:  {}
".format(results) )
print( "neighbours:  {}
".format(neighbours) )
print( "distance:  {}
".format(dist) )

plt.show()

輸出：

result:  [[1.]]

neighbours:  [[1. 1. 0.]]

distance:  [[ 29. 149. 160.]]

上面返回的是：

newcomer的標(biāo)簽，如果最近鄰算法，k=1

k-Nearest Neighbors的標(biāo)簽

從newcomer到每個(gè)最近鄰居的相應(yīng)距離

如果newcomer有大量數(shù)據(jù)，則可以將其作為數(shù)組傳遞,相應(yīng)的結(jié)果也作為矩陣獲得.

newcomers = np.random.randint(0,100,(10,2)).astype(np.float32)


plt.scatter(newcomers[:,0],newcomers[:,1],80,"g","o")

knn = cv2.ml.KNearest_create()
knn.train(trainData, cv2.ml.ROW_SAMPLE, responses)
ret, results, neighbours ,dist = knn.findNearest(newcomers, 3)

print( "result:  {}
".format(results) )
print( "neighbours:  {}
".format(neighbours) )
print( "distance:  {}
".format(dist) )

plt.show()

輸出：

result:  [[1.]
 [0.]
 [1.]
 [0.]
 [0.]
 [0.]
 [0.]
 [0.]
 [0.]
 [0.]]

neighbours:  [[0. 1. 1.]
 [0. 0. 0.]
 [1. 1. 1.]
 [0. 1. 0.]
 [1. 0. 0.]
 [0. 1. 0.]
 [0. 0. 0.]
 [0. 1. 0.]
 [0. 0. 0.]
 [0. 0. 1.]]

distance:  [[ 229.  392.  397.]
 [   4.   10.  233.]
 [  73.  146.  185.]
 [ 130.  145. 1681.]
 [  61.  100.  125.]
 [   8.   29.  169.]
 [  41.   41.  306.]
 [  85.  505.  733.]
 [ 242.  244.  409.]
 [  61.  260.  493.]]