def getErrorOfRecMethod(recType = 0):
"""Get error of all recommender.
Input:
recType - id of recommender.
Output:
recalls,precisions,f1s,maes,rmses.
"""
start_time = time.time()
songDict = persist.readSongFromFile()
allPlaylist = persist.readPlaylistFromFile_Session()
recalls = []
precisions = []
f1s = []
maes = []
rmses = []
for scale in range(10):
playlistDict = allPlaylist[scale]
if recType == const.ARIMA:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.SIMILAR:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.AVG:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
index = 0
for topN in range(1,const.TOP_N,1):
recall,precision,f1 = util.getTopNIndex(recDict,playlistDict,topN)
mae,rmse = util.getMAEandRMSE(recDict,playlistDict,songDict,topN)
if scale == 0:
recalls.append(recall)
precisions.append(precision)
f1s.append(f1)
maes.append(mae)
rmses.append(rmse)
else:
recalls[index] += recall
precisions[index] += precision
f1s[index] += f1
maes[index] += mae
rmses[index] += rmse
index += 1
#cal the avg value
recalls = [recall / 10 for recall in recalls]
precisions = [precision / 10 for precision in precisions]
f1s = [f1 / 10 for f1 in f1s]
maes = [mae / 10 for mae in maes]
rmses = [rmse / 10 for rmse in rmses]
#logging info to log
index = 0
for topN in range(1,const.TOP_N,1):
print '%d:TopN = %d:%f %f %f %f %f' % (recType,topN,recalls[index],precisions[index],f1s[index],maes[index],rmses[index])
logging.info('%d>%d:%f %f %f %f %f' % (recType,topN,recalls[index],precisions[index],f1s[index],maes[index],rmses[index]))
index += 1
end_time = time.time()
print 'Consumed:%d' % (end_time-start_time)
return recalls,precisions,f1s,maes,rmses
def errorTest():
songDict = persist.readSongFromFile()
print len(songDict)
disDict = {}
tarDict = songDict[672661981].getTopicDict()
for sid in songDict:
song = songDict[sid]
topicDict = song.getTopicDict()
dis = util.similarity(tarDict,topicDict)
disDict[sid] = dis
disList = sorted(disDict.iteritems(),key=lambda x:x[1])
for i in range(0,100):
print disList[i]
def errorTest():
songDict = persist.readSongFromFile()
print len(songDict)
disDict = {}
tarDict = songDict[672661981].getTopicDict()
for sid in songDict:
song = songDict[sid]
topicDict = song.getTopicDict()
dis = util.similarity(tarDict, topicDict)
disDict[sid] = dis
disList = sorted(disDict.iteritems(), key=lambda x: x[1])
for i in range(0, 100):
print disList[i]
def testRecMethod(recType = 0):
"""Test single recommender.
Input:
recType - id of recommender.
Output:
None.
"""
info = '############%s#############' % util.getMethodName(recType)
start_time = time.time()
songDict = persist.readSongFromFile()
allPlaylist = persist.readPlaylistFromFile()
recallTotal = 0.0
precisionTotal = 0.0
f1Total = 0.0
maeTotal = 0.0
rmseTotal = 0.0
if recType == const.KNN:
simFilename = '../txt/simMatrix.txt'
pid2Index,index2Pid,countMatrix = util.getUserSongMatrix(allPlaylist,songDict)
if os.path.exists(simFilename):
print '......'
simFile = open(simFilename,'r')
line = simFile.readline().rstrip('\n')
simMatrix = eval(line)
print len(simMatrix)
print len(simMatrix[1234])
simFile.close()
else:
print '++++++'
simMatrix = util.getUserSimMatrix(countMatrix,pid2Index)
for scale in range(10):
playlistDict = allPlaylist[scale]
playlistDict = allPlaylist[scale]
if recType == const.ARIMA:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.SIMILAR:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.AVG:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.ARIMA_SIMILAR:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.ARIMA_AVG:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.KNN:
recDict = predict.getRecDictOfUserKNN(playlistDict,songDict,scale,pid2Index,countMatrix,simMatrix)
elif recType == const.MARKOV:
recDict = predict.getRecDictOfFirstMarkov(allPlaylist,songDict,scale)
elif recType == const.MARKOV_3:
recDict = predict.getRecDictOfThreeOrderMarkov(allPlaylist,songDict,scale)
elif recType == const.PATTERN:
recDict = predict.getRecDictOfMostPattern(allPlaylist,songDict,scale)
recall,precision,f1 = util.getTopNIndex(recDict,playlistDict)
mae,rmse = util.getMAEandRMSE(recDict,playlistDict,songDict)
recallTotal += recall
precisionTotal += precision
f1Total += f1
maeTotal += mae
rmseTotal += rmse
recall = recallTotal / 10
precision = precisionTotal / 10
f1 = f1Total / 10
mae = maeTotal / 10
rmse = rmseTotal / 10
print info
logging.info(info)
print 'Recall = ',recall
logging.info('Recall = %f' % recall)
print 'Precision = ',precision
logging.info('Precision = %f' % precision)
print 'F1-Score = ',f1
logging.info('F1-Score = %f' % f1)
print 'MAE = ',mae
logging.info('MAE = %f' % mae)
print 'RMSE = ',rmse
logging.info('RMSE = %f' % rmse)
print 'Consumed: %ds' % (time.time()-start_time)
logging.info('Consumed: %ds' % (time.time()-start_time))
def getErrorOfRecMethod(recType = 0):
"""Get error of all recommender.
Input:
recType - id of recommender.
Output:
recalls,precisions,f1s,maes,rmses.
"""
start_time = time.time()
songDict = persist.readSongFromFile()
allPlaylist = persist.readPlaylistFromFile()
recalls = []
precisions = []
f1s = []
maes = []
rmses = []
if recType == const.KNN:
pid2Index,index2Pid,countMatrix = util.getUserSongMatrix(allPlaylist,songDict)
simFilename = '../txt/simMatrix.txt'
if os.path.exists(simFilename):
print '......'
simFile = open(simFilename,'r')
line = simFile.readline().rstrip('\n')
simMatrix = eval(line)
print len(simMatrix)
print len(simMatrix[1234])
simFile.close()
else:
print '++++++'
simMatrix = util.getUserSimMatrix(countMatrix,pid2Index)
for scale in range(10):
playlistDict = allPlaylist[scale]
if recType == const.ARIMA:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.SIMILAR:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.AVG:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.ARIMA_SIMILAR:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.ARIMA_AVG:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.ALL_HYBRID:
recDict = predict.getRecDict(playlistDict,songDict,recType,scale)
elif recType == const.KNN:
recDict = predict.getRecDictOfUserKNN(playlistDict,songDict,scale,pid2Index,countMatrix,simMatrix)
elif recType == const.MARKOV:
recDict = predict.getRecDictOfFirstMarkov(allPlaylist,songDict,scale)
elif recType == const.MARKOV_3:
recDict = predict.getRecDictOfThreeOrderMarkov(allPlaylist,songDict,scale)
elif recType == const.PATTERN:
recDict = predict.getRecDictOfMostPattern(allPlaylist,songDict,scale)
index = 0
for topN in range(1,const.TOP_N,1):
recall,precision,f1 = util.getTopNIndex(recDict,playlistDict,topN)
mae,rmse = util.getMAEandRMSE(recDict,playlistDict,songDict,topN)
if scale == 0:
recalls.append(recall)
precisions.append(precision)
f1s.append(f1)
maes.append(mae)
rmses.append(rmse)
else:
recalls[index] += recall
precisions[index] += precision
f1s[index] += f1
maes[index] += mae
rmses[index] += rmse
index += 1
#cal the avg value
recalls = [recall / 10 for recall in recalls]
precisions = [precision / 10 for precision in precisions]
f1s = [f1 / 10 for f1 in f1s]
maes = [mae / 10 for mae in maes]
rmses = [rmse / 10 for rmse in rmses]
#logging info to log
index = 0
for topN in range(1,const.TOP_N,1):
print '%d:TopN = %d:%f %f %f %f %f' % (recType,topN,recalls[index],precisions[index],f1s[index],maes[index],rmses[index])
logging.info('%d>%d:%f %f %f %f %f' % (recType,topN,recalls[index],precisions[index],f1s[index],maes[index],rmses[index]))
index += 1
end_time = time.time()
print 'Consumed:%d' % (end_time-start_time)
return recalls,precisions,f1s,maes,rmses
songDict,
scale)
print 'get trainingPatterns and testingPatterns...'
#print allPattern
#print len(allPattern)
#S = read("../txt/PrefixSpan.txt")
print 'get frequent pattern...'
patterns = prefixSpan(SquencePattern([], sys.maxint), allTrainingPattern, 30)
print 'Finish getting frequent pattern...'
#print_patterns(patterns)
#print len(patterns)
#count = 0
size = len(testingPatternDict)
index = 0
for pid in testingPatternDict.keys():
#print '======%d======' % pid
index += 1
print 'scale=%d > %d/%d:%d' % (scale,index,size,pid)
pattern = testingPatternDict[pid]
predictTopic = getPredictTopic(patterns,pattern)
predictTopicDict[pid] = predictTopic
#count += len(predictTopic)
#print 'avg = ',count*2.0/len(patternDict)
print 'I am out getPredictTopicDict......'
return predictTopicDict
if __name__ == "__main__":
songDict = persist.readSongFromFile()
playlistDict = persist.readPlaylistFromFile()
getPredictTopicDict(playlistDict,songDict)
def testRecMethod(recType=0):
"""Test single recommender.
Input:
recType - id of recommender.
Output:
None.
"""
info = '############%s#############' % util.getMethodName(recType)
start_time = time.time()
songDict = persist.readSongFromFile()
allPlaylist = persist.readPlaylistFromFile()
recallTotal = 0.0
precisionTotal = 0.0
f1Total = 0.0
maeTotal = 0.0
rmseTotal = 0.0
if recType == const.KNN:
simFilename = '../txt/simMatrix.txt'
pid2Index, index2Pid, countMatrix = util.getUserSongMatrix(
allPlaylist, songDict)
if os.path.exists(simFilename):
print '......'
simFile = open(simFilename, 'r')
line = simFile.readline().rstrip('\n')
simMatrix = eval(line)
print len(simMatrix)
print len(simMatrix[1234])
simFile.close()
else:
print '++++++'
simMatrix = util.getUserSimMatrix(countMatrix, pid2Index)
for scale in range(10):
playlistDict = allPlaylist[scale]
playlistDict = allPlaylist[scale]
if recType == const.ARIMA:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.SIMILAR:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.AVG:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.ARIMA_SIMILAR:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.ARIMA_AVG:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.KNN:
recDict = predict.getRecDictOfUserKNN(playlistDict, songDict,
scale, pid2Index,
countMatrix, simMatrix)
elif recType == const.MARKOV:
recDict = predict.getRecDictOfFirstMarkov(allPlaylist, songDict,
scale)
elif recType == const.MARKOV_3:
recDict = predict.getRecDictOfThreeOrderMarkov(
allPlaylist, songDict, scale)
elif recType == const.PATTERN:
recDict = predict.getRecDictOfMostPattern(allPlaylist, songDict,
scale)
recall, precision, f1 = util.getTopNIndex(recDict, playlistDict)
mae, rmse = util.getMAEandRMSE(recDict, playlistDict, songDict)
recallTotal += recall
precisionTotal += precision
f1Total += f1
maeTotal += mae
rmseTotal += rmse
recall = recallTotal / 10
precision = precisionTotal / 10
f1 = f1Total / 10
mae = maeTotal / 10
rmse = rmseTotal / 10
print info
logging.info(info)
print 'Recall = ', recall
logging.info('Recall = %f' % recall)
print 'Precision = ', precision
logging.info('Precision = %f' % precision)
print 'F1-Score = ', f1
logging.info('F1-Score = %f' % f1)
print 'MAE = ', mae
logging.info('MAE = %f' % mae)
print 'RMSE = ', rmse
logging.info('RMSE = %f' % rmse)
print 'Consumed: %ds' % (time.time() - start_time)
logging.info('Consumed: %ds' % (time.time() - start_time))
def getErrorOfRecMethod(recType=0):
"""Get error of all recommender.
Input:
recType - id of recommender.
Output:
recalls,precisions,f1s,maes,rmses.
"""
start_time = time.time()
songDict = persist.readSongFromFile()
allPlaylist = persist.readPlaylistFromFile()
recalls = []
precisions = []
f1s = []
maes = []
rmses = []
if recType == const.KNN:
pid2Index, index2Pid, countMatrix = util.getUserSongMatrix(
allPlaylist, songDict)
simFilename = '../txt/simMatrix.txt'
if os.path.exists(simFilename):
print '......'
simFile = open(simFilename, 'r')
line = simFile.readline().rstrip('\n')
simMatrix = eval(line)
print len(simMatrix)
print len(simMatrix[1234])
simFile.close()
else:
print '++++++'
simMatrix = util.getUserSimMatrix(countMatrix, pid2Index)
for scale in range(10):
playlistDict = allPlaylist[scale]
if recType == const.ARIMA:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.SIMILAR:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.AVG:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.ARIMA_SIMILAR:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.ARIMA_AVG:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.ALL_HYBRID:
recDict = predict.getRecDict(playlistDict, songDict, recType,
scale)
elif recType == const.KNN:
recDict = predict.getRecDictOfUserKNN(playlistDict, songDict,
scale, pid2Index,
countMatrix, simMatrix)
elif recType == const.MARKOV:
recDict = predict.getRecDictOfFirstMarkov(allPlaylist, songDict,
scale)
elif recType == const.MARKOV_3:
recDict = predict.getRecDictOfThreeOrderMarkov(
allPlaylist, songDict, scale)
elif recType == const.PATTERN:
recDict = predict.getRecDictOfMostPattern(allPlaylist, songDict,
scale)
index = 0
for topN in range(1, const.TOP_N, 1):
recall, precision, f1 = util.getTopNIndex(recDict, playlistDict,
topN)
mae, rmse = util.getMAEandRMSE(recDict, playlistDict, songDict,
topN)
if scale == 0:
recalls.append(recall)
precisions.append(precision)
f1s.append(f1)
maes.append(mae)
rmses.append(rmse)
else:
recalls[index] += recall
precisions[index] += precision
f1s[index] += f1
maes[index] += mae
rmses[index] += rmse
index += 1
#cal the avg value
recalls = [recall / 10 for recall in recalls]
precisions = [precision / 10 for precision in precisions]
f1s = [f1 / 10 for f1 in f1s]
maes = [mae / 10 for mae in maes]
rmses = [rmse / 10 for rmse in rmses]
#logging info to log
index = 0
for topN in range(1, const.TOP_N, 1):
print '%d:TopN = %d:%f %f %f %f %f' % (recType, topN, recalls[index],
precisions[index], f1s[index],
maes[index], rmses[index])
logging.info('%d>%d:%f %f %f %f %f' %
(recType, topN, recalls[index], precisions[index],
f1s[index], maes[index], rmses[index]))
index += 1
end_time = time.time()
print 'Consumed:%d' % (end_time - start_time)
return recalls, precisions, f1s, maes, rmses
def getResultOfArimaMethod():
"""Run MTSA with different maximum length of time series
and then return the average hit ratio, precision, f1, mae and rmse.
Input:
None.
Output:
result - result of experiments.
"""
start_time = time.time()
songDict = persist.readSongFromFile() # read total songs from file.
allPlaylist = persist.readPlaylistFromFile() # read all playlists from file.
result = []
totalRecalls = []
totalPrecisions = []
totalF1s = []
totalMaes = []
totalRmses = []
for length in range(5,51,5): # set maximum length between 5 to 50.
recalls = 0.0
precisions = 0.0
f1s = 0.0
maes = 0.0
rmses = 0.0
for scale in range(10):
playlistDict = allPlaylist[scale] # get the playlist in specific scale
recDict = predict.getRecDict(playlistDict,songDict,const.ARIMA,scale,
const.TOP_N,length) # get predicted result
#get the middle result
recall,precision,f1 = util.getTopNIndex(recDict,playlistDict,const.TOP_N)
mae,rmse = util.getMAEandRMSE(recDict,playlistDict,songDict,const.TOP_N)
#add to summary
recalls += recall
precisions += precision
f1s += f1
maes += mae
rmses += rmse
#cal the avg value
recalls = recalls / 10
precisions = precisions / 10
f1s = f1s / 10
maes = maes / 10
rmses = rmses / 10
#log info to lod file
print 'Length = %d:%f %f %f %f %f' % (length,recalls,precisions,f1s,maes,rmses)
logging.info('%d:%f %f %f %f %f' % (length,recalls,precisions,f1s,maes,rmses))
#add result to list
totalRecalls.append(recalls)
totalPrecisions.append(precisions)
totalF1s.append(f1s)
totalMaes.append(maes)
totalRmses.append(rmses)
result.append(totalRecalls)
result.append(totalPrecisions)
result.append(totalF1s)
result.append(totalMaes)
result.append(totalRmses)
end_time = time.time()
print 'Consumed:%d' % (end_time-start_time)
return result