day3 python学习笔记 knn

KNN----python

def euclidean_dist_squared(X, Xtest):
 return np.sum(X**2, axis=1)[:,None] + np.sum(Xtest**2, axis=1)[None] - 2 * np.dot(X,Xtest.T)

# -*- coding: utf-8 -*-
"""
Created on Fri May  4 09:47:05 2018

@author: sun_y
"""

import numpy as np
from scipy import stats
import utils

class my_knn:

    def __init__(self, k):
        self.k = k

    def fit(self, X, y):
        self.X = X # just memorize the trianing data
        self.y = y 

    def predict(self, Xtest):
        ''' YOUR CODE HERE FOR Q4.1.1 '''
        self.Xtest=Xtest
        t=self.Xtest.shape[0]
        D=[]
       # D = utils.euclidean_dist_squared(self.X,self.Xtest)
        D = utils.euclidean_dist_squared(self.Xtest,self.X)
        yhat =np.zeros(t)
        sort_indexD=[]
        sort_indexD=D.argsort()
        #print(np.shape(sort_indexD))
       # print(sort_indexD)
        ynear_index=[]
        ynear_index=sort_indexD[:,0:self.k]
        
        ynear=np.zeros((t,self.k))
        for i in range(t):
           for j in range(self.k):
               ynear[i,j]=np.array(self.y[ynear_index[i,j]])
           yhat[i]=utils.mode(ynear[i,:])
        return yhat  

【定义数组】

     c=np.array([[1,2],[3,4]])

【numpy.shape】 x.shape[0]=2    np.shape(x)=(2,2)

【numpy.sort】   x.sort()     x.argsort() 

***************************************************************************************************************************

【load CSV file】

                          

import pandas as pd                        

 # pandas 模块：点击打开链接

locationNum=5&fps=1

import os                                               # os 模块：点击打开链接

df = pd.read_csv(os.path.join('E:\\assignment 2018 with code\\p2o1b_a1-master\\p2o1b_a1-master\\data','fluTrends.csv'))

import numpy as np
from sklearn.cross_validation import train_test_split
from sklearn.datasets import load_iris
from my_knn import my_knn
iris = load_iris()
n, d = iris.data.shape
"""
print(iris.keys())

print((n_samples, n_features))
print(iris.data[0])
print(iris.target.shape)
print(iris.target)
print(iris.target_names)
print("feature_names:",iris.feature_names)
"""
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target, test_size=0.1)

KValue=4
model=my_knn(k=KValue)
model.fit(X_train,y_train)
y_pred = model.predict(X_train)
train_error = np.mean(y_pred != y_train)
print("train error is %f" % train_error)
y_pred_test=model.predict(X_test)
test_error=np.mean(y_pred_test != y_test)
print("test error is %f" % test_error)

【测试结果：】

k=4时

train error is 0.029630
test error is 0.066667

***************************************分割线******************************************************************

【python 自带KNN】

import numpy as np    
from sklearn import neighbors   
model2 = neighbors.KNeighborsClassifier() #取得knn分类器  
model2.fit(X_train,y_train)
y_pred = model2.predict(X_train)
train_error = np.mean(y_pred != y_train)
print("train error is %f" % train_error)
y_pred_test=model2.predict(X_test)
test_error=np.mean(y_pred_test != y_test)
print("test error is %f" % test_error)

train error is 0.022222
test error is 0.066667

测试结果一致

【返回 true 和 false】

return True # 注意大写