#-*- encoding:utf-8 -*- import numpy as np import pandas as pd def test(): # header=0,表示文件第0行为列索引 # index_col=0,表示文件第0列为行索引 # typeTable=pd.read_csv('./data/type.csv',header=None) userRatingTable=pd.read_csv('./data/RatingTable.csv',header=None) userRatingTable.index=range(0,943)#行号 userRatingTable.columns=range(0,1682)#列号 i=0#i表示文件号 # print(userTable[1])#取第1列,#注意::这里前后颠倒了,表示矩阵中用户10对项目4的评分,即评分为4 for j in range(943): # print(userRatingTable.ix[942])#取第一行 temp=userRatingTable.ix[j] userTable=pd.read_csv('./data/preprefe_%s.csv'%str(i),header=0,index_col=0) userTable.index=range(0,1682) userTable.columns=range(0,19) simMat=[] Userscore=[] count=0 for k in range(1682): # print(temp[k]) # print(type(temp[k])) if temp[k]!=0: # print(temp[k]) count=count+1 # print(temp[k]) # print(userTable.ix[k]) Userscore.append(temp[k]) simMat.append(userTable.ix[k]) # print(simMat) data_to_excel=pd.DataFrame(simMat,index=np.arange(0,count),columns=np.arange(0,19)) data_to_excel.to_csv('./data/data1/preprefeggg_%s.csv'%str(i)) print(Userscore,count) simMat=[] Userscore=[] count=0 # if i==3: # exit() i=i+1 if __name__=='__main__': test()