def dumpPickles(id):
"""
Persist memory intensive data structures, item similarities.
"""
print 'Pickle id %s' % (id)
foldername = '../kfold_cluster_' + id + '/'
print foldername
cluster = algorithm.createPrefs(folder=foldername)
for c in range(8):
prefs = cluster[c]
itemsim = algorithm.calculateSimilarItems(prefs, n=50)
dumpname = "../pickle/" + id + "/oitemsim_" + str(c) + ".pickle"
pickle.dump(itemsim, open(dumpname, "wb"))
def dumpPickles(id):
"""
Persist memory intensive data structures, item similarities.
"""
print 'Pickle id %s'%(id)
foldername = '../kfold_cluster_' + id + '/'
print foldername
cluster = algorithm.createPrefs(folder=foldername)
for c in range(8):
prefs = cluster[c]
itemsim = algorithm.calculateSimilarItems(prefs, n=50)
dumpname = "../pickle/" +id + "/oitemsim_" + str(c) + ".pickle"
pickle.dump(itemsim, open(dumpname, "wb"))
sparsity.caseone_doAverageFilling(mod_prefs, deleted_users)
sparsity.saveDenseMatrix(mod_prefs,filename='..\\dataset\\cluster_testing10.tab')
genetic_som.MAX_CLUSTERS = 10
genetic_som.MIN_CLUSTERS = 4
genetic_som.setFilename= '../dataset/cluster_testing10.tab'
genetic_som.genetic(pop=20,max_evals=100,elites=4)
#persist clusters onto files. UNVERIFIED
genetic_som.persistClusters(genetic_som.getChromosomeList(genetic_som.final_pop),
filename='../dataset/cluster_testing10.tab', folder='cluster',clustername='cluster_testing')
"""
cluster = algorithm.createPrefs(folder='../cluster/')
#get cluster mean centroids.
cluster_centroids = getClusterCentroids(cluster)
for user in deleted_users:
user_dict, removed_items = getRemovedPrefs(original_prefs[str(user)],
10)
b = getBestCluster(user_dict)
print "User %s Cluster %s" % (user, b)
print "Prefs after removal of items: "
print user_dict
new_prefs = {}
new_prefs = cluster[int(b)]
genetic_som.MAX_CLUSTERS = 10
genetic_som.MIN_CLUSTERS = 4
genetic_som.setFilename= '../dataset/three_dense_matrix.tab'
genetic_som.genetic(pop=20,max_evals=100,elites=4)
#persist clusters onto files. UNVERIFIED
genetic_som.persistClusters(genetic_som.getChromosomeList(genetic_som.final_pop),
filename='../dataset/three_dense_matrix.tab', folder='three',clustername='threecluster')
"""
cluster = algorithm.createPrefs(folder='../five/')
testingPrefs = getTestPrefs(original_prefs, n=5)
#use the modified prefs here.
ostart = time.time()
oitemsim = original.calculateSimilarItems(testingPrefs, n=50)
ostop = time.time()
#keep track of cluster prediction details
colist = []
calist = []
coMAE = []
caMAE = []
for cid in range(8):
# training[str(key + 1)] = original_prefs[str(key+1)]
"""
sparsity.clustering_doFillUsingCBR(training)
sparsity.clustering_doAverageFilling(training)
sparsity.saveDenseMatrix(training,filename='..\\kfold\\kfold_testing_five.tab')
genetic_som.persistClusters([0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1],
filename='..\\kfold\\kfold_testing_five.tab', folder='kfold_cluster_five',clustername='kfold_cluster_testing_five')
"""
cid = 'four'
cluster = algorithm.createPrefs(folder='../kfold_cluster_' + cid + '/')
#get cluster mean centroids.
cluster_centroids = getClusterCentroids(cluster)
for user in validation:
user_dict, removed_items = getRemovedPrefs(validation[user], 10)
b = getBestCluster(user_dict)
print "User %s Cluster %s" % (user, b)
#print "Prefs after removal of items: "
#print user_dict
# new prefs for calculation after removing 10 items from validation set of the user.
new_prefs = {}
new_prefs = cluster[int(b)]
"""
sparsity.clustering_doFillUsingCBR(training)
sparsity.clustering_doAverageFilling(training)
sparsity.saveDenseMatrix(training,filename='..\\kfold\\kfold_testing_five.tab')
genetic_som.persistClusters([0, 0, 0, 1, 1, 1, 0, 0, 0, 1, 1],
filename='..\\kfold\\kfold_testing_five.tab', folder='kfold_cluster_five',clustername='kfold_cluster_testing_five')
"""
cid = 'four'
cluster = algorithm.createPrefs(folder='../kfold_cluster_' + cid +'/')
#get cluster mean centroids.
cluster_centroids = getClusterCentroids(cluster)
for user in validation:
user_dict, removed_items = getRemovedPrefs(validation[user], 10)
b = getBestCluster(user_dict)
print "User %s Cluster %s"%(user, b)
#print "Prefs after removal of items: "
#print user_dict
# new prefs for calculation after removing 10 items from validation set of the user.
new_prefs = {}
sparsity.caseone_doAverageFilling(mod_prefs, deleted_users)
sparsity.saveDenseMatrix(mod_prefs,filename='..\\dataset\\cluster_testing10.tab')
genetic_som.MAX_CLUSTERS = 10
genetic_som.MIN_CLUSTERS = 4
genetic_som.setFilename= '../dataset/cluster_testing10.tab'
genetic_som.genetic(pop=20,max_evals=100,elites=4)
#persist clusters onto files. UNVERIFIED
genetic_som.persistClusters(genetic_som.getChromosomeList(genetic_som.final_pop),
filename='../dataset/cluster_testing10.tab', folder='cluster',clustername='cluster_testing')
"""
cluster = algorithm.createPrefs(folder='../cluster/')
#get cluster mean centroids.
cluster_centroids = getClusterCentroids(cluster)
for user in deleted_users:
user_dict, removed_items = getRemovedPrefs(original_prefs[str(user)], 10)
b = getBestCluster(user_dict)
print "User %s Cluster %s"%(user, b)
print "Prefs after removal of items: "
print user_dict
new_prefs = {}
new_prefs = cluster[int(b)]
new_prefs[str(user)] = user_dict