from sklearn.neighbors import kneighborsclassifier
from math import sqrt
import os
import pandas as pd
import numpy as np
from collections import counter
path=os.getcwd()+
'\\'
+'8c859681-527c-4d9b-a98a-f00e6169931f_.xls'
x1=pd.read_excel(path)
x1=np.array(x1)
x=x1[:,
:2]row_data_x=x
row_data_y=
for i in
range
(len
(x))
:if i<=
len(x)//2
:0)else:1
)row_data_y=np.array(row_data_y)
#訓練集
x_train=np.array(row_data_x)
y_train=np.array(row_data_y)
x=x1[:,
:2]knn_classfier=kneighborsclassifier(n_neighbors=6)
knn_classfier.fit(x_train,y_train)
x_predict=x
y_predict=knn_classfier.predict(x_predict)
class
knnclassfier
:def
__init__
(self,k)
:'''初始化knn分類器'''
assert k>=1,
"k must be valid"
self.k=k
self._x_train=
none
self._y_train=
none
deffit
(self,x_train,y_train)
:'''根據訓練資料集x_train和y_train訓練knn分類器'''
assert x_train.shape[0]
==y_train.shape[0]
,\ "the size of x_train must be equal to the size of train"
assert self.k<=x_train.shape[0]
,\ "the size of x_train must be at least l."
self._x_train=x_train
self._y_train=y_train
return self
defpredict
(self,x_predict)
:'''給定帶**資料集x)predict,返回表示x_predict的結果向量'''
assert self._x_train is
notnone
and self._y_train is
notnone
,\ "must fit before predict"
assert x_predict.shape[1]
==self._x_train.shape[1]
,\ "the feature number of x_predict must be equal to x_train"
y_predict=
[self._predict(x)
for x in x_predict]
return np.array(y_predict)
def_predict
(self,x)
:'''給定單個帶**資料x,返回x的**結果值'''
assert x.shape[0]
==self._x_train.shape[1]
, \ "must feature number of x must be equal to x_train"
distances=
[sqrt(np.
sum(
(x_train-x)**2
))for x_train in self._x_train]
nearest=np.argsort(distances)
topk_y=
[self._y_train[i]
for i in nearest[
:self.k]
] votes=counter(topk_y)
return votes.most_common(1)
[0][
0]def__repr__
(self)
:return
"knn(k=%d"
%self.k
if __name__==
'__main__'
: knn_clf= knnclassfier(k=6)
knn_clf.fit(x_train,y_train)
x_predict=x
y_predict=knn_clf.predict(x_predict)
print
(y_predict)
深度學習 學習筆記 之最鄰近演算法KNN（2）
1 cover和hart在1968年提出了最初的鄰近演算法 2 knn 演算法的核心思想是如果乙個樣本在特徵空間中的k個最相鄰的樣本中的大多數屬於某乙個類別，則該樣本也屬於這個類別，並具有這個類別上樣本的特性。3 類別決策時，只與極少量的相鄰樣本有關。4 距離的測量方法 euclidean dist...
機器學習之Knn演算法2
寫了knn演算法以後，再寫乙個二，一的連線 coding utf 8 import math import random import operator import csv import numpy as np defloaddataset filename,split,trainingset t...
機器學習 KNN
posted on 2015 05 21 20 34 in ml 0 comment views 3195 原始碼 決策樹和基於規則的分類器都是積極學習方法 eager learner 的例子，因為一旦訓練資料可用，他們就開始學習從輸入屬性到類標號的對映模型。乙個相反的策略是推遲對訓練資料的建模，直...