test[user] = userSequence[user][partition:]
for user in train:
print user , train[user]
userData_trainset.write(json.dumps({user:train[user]})+"\n")
for user in test:
userData_testset.write(json.dumps({user:test[user]})+"\n")
userData_trainset.close()
userData_testset.close()
userData.close()
"""
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userProfiles -reduceDimensions -userSimilarity=6040 -formClusters=0.8")
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userSimilarity=6040 -formClusters=0.8")
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", buildGraph=False, userProfiles=False, generateSequence=None, reduceDimensions=False, userSimilarity=6040, formClusters=None, uid=None)
for user in userSequence:
print "generating data for user " + user
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", uid=user)
os.system("mv movielens_" + user + "_contentReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_collabReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_combinedReco.pickle " + str(percent))
os.system("mv movielens_userProfiles.pickle " + str(percent))
os.system("mv movielens_attributeRelativeImportance.pickle " + str(percent))
os.system("mv movielens_userSimilarity.pickle " + str(percent))
os.system("mv movielens_clusters.pickle " + str(percent))
print "processing user ", user
partition = int(math.ceil(len(userSequence[user]) * percent / 100.0))
train[user] = userSequence[user][:partition]
test[user] = userSequence[user][partition:]
for user in train:
print user, train[user]
userData_trainset.write(json.dumps({user: train[user]}) + "\n")
for user in test:
userData_testset.write(json.dumps({user: test[user]}) + "\n")
userData_trainset.close()
userData_testset.close()
userData.close()
"""
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userProfiles -reduceDimensions -userSimilarity=6040 -formClusters=0.8")
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userSimilarity=6040 -formClusters=0.8")
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", buildGraph=False, userProfiles=False, generateSequence=None, reduceDimensions=False, userSimilarity=6040, formClusters=None, uid=None)
for user in userSequence:
print "generating data for user " + user
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", uid=user)
os.system("mv movielens_" + user + "_contentReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_collabReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_combinedReco.pickle " + str(percent))
os.system("mv movielens_userProfiles.pickle " + str(percent))
os.system("mv movielens_attributeRelativeImportance.pickle " + str(percent))
os.system("mv movielens_userSimilarity.pickle " + str(percent))
os.system("mv movielens_clusters.pickle " + str(percent))
for user in userSequence:
print "processing user ", user
partition = int(math.ceil(len(userSequence[user])*percent/100.0))
train[user] = userSequence[user][:partition]
test[user] = userSequence[user][partition:]
userData_trainset.write(json.dumps(train))
userData_testset.write(json.dumps(test))
userData_trainset.close()
userData_testset.close()
userData.close()
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userProfiles -reduceDimensions -userSimilarity=6040 -formClusters=0.8")
#os.system("python modules.py --db=movielens --usageData=movielens_userData.json -userSimilarity=6040 -formClusters=0.8")
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", buildGraph=False, userProfiles=False, generateSequence=None, reduceDimensions=False, userSimilarity=6040, formClusters=None, uid=None)
for user in userSequence:
print "generating data for user " + user
modules.mainImport(db="movielens", usageData="movielens_userData_trainset.json", uid=user)
os.system("mv movielens_" + user + "_contentReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_collabReco.pickle " + str(percent))
os.system("mv movielens_" + user + "_combinedReco.pickle " + str(percent))
os.system("mv movielens_userProfiles.pickle " + str(percent))
os.system("mv movielens_attributeRelativeImportance.pickle " + str(percent))
os.system("mv movielens_userSimilarity.pickle " + str(percent))
os.system("mv movielens_clusters.pickle " + str(percent))
os.system("mv movielens_userData_trainset.json " + str(percent))
os.system("mv movielens_userData_testset.json " + str(percent))
