Python pearsSim Exemples

Exemple #1

Fichier : test.py Projet : wang4yu6peng13/MachineLearningInAction

import svdRec
from numpy import *

Data = svdRec.loadExData()
U, Sigma, VT = linalg.svd(Data)
print Sigma

Sig3 = mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[2]]])
print U[:,:3] * Sig3 * VT[:3,:]

myMat = mat(svdRec.loadExData())
print svdRec.ecludSim(myMat[:,0], myMat[:,4])
print svdRec.ecludSim(myMat[:,0], myMat[:,1])
print svdRec.pearsSim(myMat[:,0], myMat[:,4])
print svdRec.pearsSim(myMat[:,0], myMat[:,1])
print svdRec.cosSim(myMat[:,0], myMat[:,4])
print svdRec.cosSim(myMat[:,0], myMat[:,1])

myMat = mat(svdRec.loadExData1())
print myMat
print svdRec.recommend(myMat, 2)
print svdRec.recommend(myMat, 2, simMeas=svdRec.ecludSim)
print svdRec.recommend(myMat, 2, simMeas=svdRec.pearsSim)

from numpy import linalg as la
U, Sigma, VT = la.svd(mat(svdRec.loadExData2()))
print Sigma
Sig2 = Sigma**2
print sum(Sig2)
print sum(Sig2)*0.9
print sum(Sig2[:2])

Exemple #2

def test2():
    myMat = mat(svdRec.loadExData())
    print myMat
    print svdRec.ecludSim(myMat[:, 0], myMat[:, 4])
    print svdRec.cosSim(mat([[4], [2]]), mat([[1], [2]]))
    print svdRec.pearsSim(myMat[:, 0], myMat[:, 4])

Exemple #3

Fichier : test_sim.py Projet : wcleithinking/machine_learning_in_action

from numpy import *
import svdRec

myMat = mat(svdRec.loadExData())
test1 = svdRec.ecludSim(myMat[:, 0], myMat[:, 4])
test2 = svdRec.ecludSim(myMat[:, 0], myMat[:, 0])
print(test1, test2)
test1 = svdRec.cosSim(myMat[:, 0], myMat[:, 4])
test2 = svdRec.cosSim(myMat[:, 0], myMat[:, 0])
print(test1, test2)
test1 = svdRec.pearsSim(myMat[:, 0], myMat[:, 4])
test2 = svdRec.pearsSim(myMat[:, 0], myMat[:, 0])
print(test1, test2)

result = svdRec.recommend(myMat, 2)
print(result)
result = svdRec.recommend(myMat, 2, simMeans=svdRec.ecludSim)
print(result)
result = svdRec.recommend(myMat, 2, simMeans=svdRec.pearsSim)
print(result)

Exemple #4

# SVD implement

from numpy import *

U, Sigma, VT = linalg.svd([[1, 1], [7, 7]])
#print (U,Sigma,VT)

import svdRec

Data = svdRec.loadExData()
U, Sigma, VT = linalg.svd(Data)
print(Sigma)  # 返回对角线的元素节省空间
print(Sigma.shape)

Sig3 = mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[1]]])

rec_image = U[:, :3] * Sig3 * VT[:3, :]
print(rec_image)

myMat = mat(svdRec.loadExData())
dis = svdRec.ecludSim(myMat[:, 0], myMat[:, 4])
print(dis)

dis = svdRec.cosSim(myMat[:, 0], myMat[:, 4])
print dis

dis = svdRec.pearsSim(myMat[:, 0], myMat[:, 4])
print dis

svdRec.recommend(myMat, 2)

Exemple #5

Fichier : 14.py Projet : niumeng07/ML

print("VT=",VT)

Sig3=mat([[Sigma[0],0,0],[0,Sigma[1],0],[0,0,Sigma[2]]])
Sig5=mat([[Sigma[0],0,0,0,0],[0,Sigma[1],0,0,0],[0,0,Sigma[2],0,0],[0,0,0,Sigma[3],0],[0,0,0,0,Sigma[4]]])
DataWithSig3=U[:,:3]*Sig3*VT[:3,:]
#DataWithSig5=U[:,:]*Sig5*VT[:,:]
print("Data Sigma3 ReConstruct:\n",U[:,:3]*Sig3*VT[:3,:])
#print("Data Sigma5 ReConstruct:\n",DataWithSig5)


myMat=mat(svdRec.loadExData())
print(svdRec.ecludSim(myMat[:,0],myMat[:,4]))
print(svdRec.ecludSim(myMat[:,0],myMat[:,0]))
print(svdRec.cosSim(myMat[:,0],myMat[:,4]))
print(svdRec.cosSim(myMat[:,0],myMat[:,0]))
print(svdRec.pearsSim(myMat[:,0],myMat[:,4]))
print(svdRec.pearsSim(myMat[:,0],myMat[:,0]))

myMat=mat([[4,4,0,2,2],[4,0,0,3,3],[4,0,0,1,1],[1,1,1,2,0],
    [2,2,2,0,0],[1,1,1,0,0],[5,5,5,0,0]])
print(myMat)

recommendResult=svdRec.recommend(myMat,2)
recommendResultWithEclud=svdRec.recommend(myMat,2,simMeas=svdRec.ecludSim)
recommendResultWithPears=svdRec.recommend(myMat,2,simMeas=svdRec.pearsSim)

print(recommendResult)
print(recommendResultWithEclud)
print(recommendResultWithPears)

U,Sigma,VT=linalg.svd(mat(svdRec.loadExData2()))

Exemple #6

data = svdRec.loadExData()
U, Sigma, VT = np.linalg.svd(data)  # 在一个稍微大点的数据集上看看效果
Sigma
# 前三个数值明显大于后两个，数量及上差太多了~
# 所以呢？原数据集可以用Data(m*n) = U(m*3) * Σ(3*3) * V^T(3*n) 来近似？
Sig3 = np.mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[2]]])
U[:, :3] * Sig3 * VT[:3, :]  # 之后就用这个了？

reload(svdRec)
myMat = np.mat(svdRec.loadExData())
# 以下都是查看列向量的相似度
svdRec.eculidSim(myMat[:, 0], myMat[:, 4])  # 欧氏距离
svdRec.eculidSim(myMat[:, 0], myMat[:, 0])
svdRec.cosSim(myMat[:, 0], myMat[:, 4])  # 余弦相似度
svdRec.cosSim(myMat[:, 0], myMat[:, 0])
svdRec.pearsSim(myMat[:, 0], myMat[:, 4])  # 皮尔逊相关系数
svdRec.pearsSim(myMat[:, 0], myMat[:, 0])
'''
    *关于是用列向量还是行向量的问题*
    如果m>>n，也就是样本数大于特征数，则用列向量进行计算相似度，因为减少计算量
    如果m<<n，也就是样本数小于特征数，则用行向量进行计算相似度，也是节省
'''

reload(svdRec)
myMat = np.mat(svdRec.loadExData())
myMat[0, 1] = myMat[0, 0] = myMat[1, 0] = myMat[2, 0] = 4
myMat[3, 3] = 2
myMat
svdRec.recommend(myMat, 2)  # 用户2，未评分的商品的评分情况，元组中第一个是商品ID，第二个是评分
svdRec.recommend(myMat, 2, simMeas=svdRec.eculidSim)  # 换一下相关度计算方法
svdRec.recommend(myMat, 2, simMeas=svdRec.pearsSim)

Exemple #7

import svdRec
from numpy import *

Data = svdRec.loadExData()
U, Sigma, VT = linalg.svd(Data)
print Sigma

Sig3 = mat([[Sigma[0], 0, 0], [0, Sigma[1], 0], [0, 0, Sigma[2]]])
print U[:, :3] * Sig3 * VT[:3, :]

myMat = mat(svdRec.loadExData())
print svdRec.ecludSim(myMat[:, 0], myMat[:, 4])
print svdRec.ecludSim(myMat[:, 0], myMat[:, 1])
print svdRec.pearsSim(myMat[:, 0], myMat[:, 4])
print svdRec.pearsSim(myMat[:, 0], myMat[:, 1])
print svdRec.cosSim(myMat[:, 0], myMat[:, 4])
print svdRec.cosSim(myMat[:, 0], myMat[:, 1])

myMat = mat(svdRec.loadExData1())
print myMat
print svdRec.recommend(myMat, 2)
print svdRec.recommend(myMat, 2, simMeas=svdRec.ecludSim)
print svdRec.recommend(myMat, 2, simMeas=svdRec.pearsSim)

from numpy import linalg as la
U, Sigma, VT = la.svd(mat(svdRec.loadExData2()))
print Sigma
Sig2 = Sigma**2
print sum(Sig2)
print sum(Sig2) * 0.9
print sum(Sig2[:2])