def main():
testbase = 'ub'
dataset_name = "u1m.data"
f = open('../testres.txt', 'a+')
k_val = [5]
per = [30]
#per = [1.4]
#metric = 'pr'
metric = 'mae'
#k_val = [5 9, 13, 17, 21]
test_name = "New User-based Test on: " + dataset_name + ' '
if item_based:
testbase = 'ib'
test_name = "New Item-based test on: " + dataset_name + ' '
iterate = product(per, k_val)
for per, k in iterate:
f.write('\n')
timestamp = strftime("%Y-%m-%d %H:%M:%S", gmtime())
f.write(test_name + timestamp + ' --> ')
e = Evaluater(dataset_name,
rec_type=testbase,
k=k,
test_percentage=per,
eval_metric=metric)
f.write(
str([per, k, e.eval_metric, e.sim_method.func_name]) + ' Error: ')
f.write(str(e.evaluate()))
f.close()
def iterate_epoch(self,
model,
lr,
epoch,
weight_decay=0,
warmup=0,
lr_decay_rate=1,
lr_decay_every=10,
eval_every=5,
early_stop=False):
eval_model = Evaluater(self.data_dir, model_name=self.model_name)
#es = EarlyStop(self.data_dir[0:-6] + 'early_stopping/', self.model_name, patience=6)
es = EarlyStop('../data_beauty_2core_es/early_stopping/',
self.model_name,
patience=6)
plot_loss_list = []
plot_score_list = []
for i in range(epoch):
plot_loss_list.extend(
self.iterate_train(model,
lr=lr,
weight_decay=weight_decay,
print_every=10000))
# early stop
if early_stop:
pre_model = es.early_stop(model)
if pre_model:
print('Early Stop eposh: {}'.format(i + 1))
return eval_model.topn_map(pre_model)
# lrスケジューリング
if i > warmup:
if (i - warmup) % lr_decay_every == 0:
lr = lr * lr_decay_rate
if (i + 1) % eval_every == 0:
#score = eval_model.topn_precision(model)
#print('epoch: {} precision: {}'.format(i, score))
score = eval_model.topn_map(model)
print('epoch: {} map: {}'.format(i, score))
plot_score_list.append(score)
#self._plot(plot_loss_list)
#self._plot(plot_score_list)
#return eval_model.topn_precision(model)
return eval_model.topn_map(model)
def cal_loss_redundancy(resource, availability, replication, lease):
replication_count = 0
data_loss = 0
data_count = 100
data_iterated = 0
lease_period = int(lease)
desired_availability = float(availability)
iteration = 0
bad_iteration = 0
if replication == 'random1copy' or replication == 'min1copy':
copy_number = 1
elif replication == 'random2copy' or replication == 'min2copy':
copy_number = 2
elif replication == 'random3copy' or replication == 'min3copy':
copy_number = 3
while iteration < 30:
quantile = int(range_minute * 0.05)
time_point = random.randint(start_minute + quantile,
end_minute - quantile)
job_count = len(interval_tree[time_point])
# evaluate sizes of data set and job set
if job_count < data_count * 3 * copy_number:
# print "Error : job set is less than 3 times of data set"
bad_iteration += 1
continue
availability_dict = dict()
for data_index in range(data_count):
data_name = "data" + str(data_index)
availability_dict[data_name] = desired_availability
match_job_dict = scheduler.schedule(time_point, lease_period,
availability_dict)
if not match_job_dict:
# print "Error : match_job_dict is none"
bad_iteration += 1
continue
for job in match_job_dict:
replication_count += len(match_job_dict[job])
evaluater = Evaluater(interval_tree, job_dict)
data_iterated = data_iterated + data_count
data_loss += evaluater.evaluate(time_point + lease_period,
match_job_dict)
iteration += 1
data_loss_rate = float(data_loss) / float(data_iterated)
redundancy_rate = float(replication_count) / float(data_iterated)
print "data loss rate : ", data_loss_rate
print "redundancy : ", redundancy_rate
print "bad iteration : ", bad_iteration
return (data_loss_rate, redundancy_rate)
def __init__(self, data_dir):
# 本当はAmazonDatasetクラスを渡した方が速いが、
self.evaluater = Evaluater(data_dir)
self.dataset = AmazonDataset(data_dir, model_name='TransE')
edges = [[r[0], r[1]] for r in self.dataset.triplet_df.values]
# user-itemとitem-userどちらの辺も追加
for r in self.dataset.triplet_df.values:
if r[2] == 0:
edges.append([r[1], r[0]])
# load network
self.G = nx.DiGraph()
self.G.add_nodes_from(
[i for i in range(len(self.dataset.entity_list))])
self.G.add_edges_from(edges)
def objective(trial):
start = time.time()
# ハイパラ読み込み
# gamma = trial.suggest_loguniform('gamma', 1e-6, 1e-3)
# lin_model = trial.suggest_categorical('lin_model', ['lasso', 'elastic'])
alpha = trial.suggest_uniform('alpha', 0, 1)
beta = trial.suggest_uniform('beta', 0, 0.5)
data_dirs = [
'../' + data_path + '/valid1/', '../' + data_path + '/valid2/'
]
score_sum = 0
for data_dir in data_dirs:
# dataload
dataset = AmazonDataset(data_dir)
# laod model
#slim = train_SLIM(data_dir, load=True)
sim_mat = load_sim_mat('sim_mat' + data_dir[-2] + '.csr',
len(dataset.user_list), len(dataset.item_list))
edges = [[r[0], r[1]] for r in dataset.triplet_df.values]
# user-itemとitem-userどちらの辺も追加
for r in dataset.triplet_df.values:
if r[2] == 0:
edges.append([r[1], r[0]])
# load network
G = nx.DiGraph()
G.add_nodes_from([i for i in range(len(dataset.entity_list))])
G.add_edges_from(edges)
evaluater = Evaluater(data_dir)
#ranking_mat = get_ranking_mat(G, slim, alpha, beta, dataset)
ranking_mat = get_ranking_mat(G, sim_mat, alpha, beta, dataset)
#score = evaluater.topn_map(ranking_mat)
score = evaluater.topn_precision(ranking_mat)
score_sum += score
mi, sec = time_since(time.time() - start)
print('{}m{}s'.format(mi, sec))
return -1 * score_sum / 2
def objective(trial):
start = time.time()
# hyper parameter
#gamma = trial.suggest_loguniform('gamma', 1e-6, 1e-3)
#lin_model = trial.suggest_categorical('lin_model', ['lasso', 'elastic'])
#slim = train_SLIM(lin_model, gamma)
alpha = trial.suggest_uniform('alpha', 0, 0.5)
beta = trial.suggest_uniform('beta', 0, 0.5)
gamma1 = trial.suggest_uniform('gamma1', 0, 1)
gamma2 = trial.suggest_uniform('gamma2', 0, 1)
gamma3 = trial.suggest_uniform('gamma3', 0, 1)
gamma = [gamma1, gamma2, gamma3]
data_dir = ['../data_luxury_5core/valid1', '../data_luxury_5core/valid2']
score_sum = 0
for i in range(len(data_dir)):
# dataload
dataset = AmazonDataset(data_dir[i], model_name='TransE')
edges = [[r[0], r[1]] for r in dataset.triplet_df.values]
# user-itemとitem-userどちらの辺も追加
for r in dataset.triplet_df.values:
if r[2] == 0:
edges.append([r[1], r[0]])
#user_items_test_dict = pickle.load(open('./data/user_items_test_dict.pickle', 'rb'))
# load network
G = nx.DiGraph()
G.add_nodes_from([i for i in range(len(dataset.entity_list))])
G.add_edges_from(edges)
ranking_mat = get_ranking_mat(G, dataset, model[i], gamma, alpha, beta)
#score = topn_precision(ranking_mat, user_items_test_dict)
evaluater = Evaluater(data_dir[i])
score = evaluater.topn_map(ranking_mat)
score_sum += score
mi, sec = time_since(time.time() - start)
print('{}m{}sec'.format(mi, sec))
return -1 * score_sum / 2
def objective(trial):
start = time.time()
# hyper parameter
alpha = trial.suggest_uniform('alpha', 0, 0.5)
beta = trial.suggest_uniform('beta', 0, 0.5)
gamma1 = trial.suggest_uniform('gamma1', 0, 1)
gamma2 = trial.suggest_uniform('gamma2', 0, 1)
gamma3 = trial.suggest_uniform('gamma3', 0, 1)
gamma = [gamma1, gamma2, gamma3]
data_dir = ['../' + data_path + '/valid1', '../' + data_path + '/valid2']
score_sum = 0
for i in range(len(data_dir)):
# dataload
dataset = AmazonDataset(data_dir[i], model_name='SparseTransE')
# load network
edges = [[r[0], r[1]] for r in dataset.triplet_df.values]
# user-itemとitem-userどちらの辺も追加
for r in dataset.triplet_df.values:
if r[2] == 0:
edges.append([r[1], r[0]])
G = nx.DiGraph()
G.add_nodes_from([i for i in range(len(dataset.entity_list))])
G.add_edges_from(edges)
ranking_mat = get_ranking_mat(G, dataset, model[i], gamma, alpha, beta)
#score = topn_precision(ranking_mat, user_items_test_dict)
evaluater = Evaluater(data_dir[i])
score = evaluater.topn_map(ranking_mat)
score_sum += score
mi, sec = time_since(time.time() - start)
print('{}m{}sec'.format(mi, sec))
return -1 * score_sum / 2
# load param
params = load_params()
alpha = params['alpha']
beta = params['beta']
gamma1 = params['gamma1']
gamma2 = params['gamma2']
gamma3 = params['gamma3']
gamma = [gamma1, gamma2, gamma3]
# dataload
dataset = AmazonDataset(data_dir, model_name='TransE')
# load network
edges = [[r[0], r[1]] for r in dataset.triplet_df.values]
# user-itemとitem-userどちらの辺も追加
for r in dataset.triplet_df.values:
if r[2] == 0:
edges.append([r[1], r[0]])
G = nx.DiGraph()
G.add_nodes_from([i for i in range(len(dataset.entity_list))])
G.add_edges_from(edges)
ranking_mat = get_ranking_mat(G, dataset, model, gamma, alpha, beta)
evaluater = Evaluater(data_dir)
score = evaluater.topn_map(ranking_mat)
mi, sec = time_since(time.time() - start)
print('{}m{}sec'.format(mi, sec))
np.savetxt('score_transe3.txt', np.array([score]))
def main(p):
start = time.time()
# 选择文件名以'json.gz'结尾的记录
file_name_list = filter(lambda x: x.endswith('json.gz'), os.listdir(p))
# TODO 添加文件是否是24个的判断(glob模块)
for file_name in file_name_list:
with open(os.path.join(p, file_name), 'r') as f:
raw_json_file = gzip.GzipFile(fileobj=f)
record_cleaner = Cleaner()
record_grouper = Grouper(db)
record_normalizer = Normalizer(db)
mongo_helper = MongoHelper(db)
counter = ActorCounter()
evaluater = Evaluater()
# 数据清洗
record_cleaner.set_dirty_data(raw_json_file)
record_cleaner.clean()
clean_record = record_cleaner.get_clean_data()
log.log('clean record %s' % len(clean_record))
# 数据处理
# 分组
record_grouper.set_records(clean_record)
record_grouper.group()
record_actor_exist = record_grouper.get_group_1()
record_actor_new = record_grouper.get_group_2()
log.log('record_actor_exist: %s' % len(record_actor_exist))
log.log('record_actor_new: %s' % len(record_actor_new))
# 处理记录的actor已存在的记录
log.log('Begin processing actor-exist records...')
# 只需要删掉记录的actor_attrs即可
for record in record_actor_exist:
del record['actor_attributes']
log.log('Finished.')
# 处理记录的actor不存在的记录
record_normalizer.set_records(record_actor_new)
record_normalizer.normalize()
record_actor_new = record_normalizer.get_record_actor_new()
new_actors = record_normalizer.get_new_actors()
# 把本地的今日新增的Actor更新到数据库
actors = new_actors.values()
mongo_helper.insert_new_actors(actors)
# 对新增的Actor, 改变Redis中相应的计数
counter.count_actor_list(actors)
# 计算每条记录的val
evaluater.set_records(record_actor_exist)
evaluater.evaluate()
val_actor_exist = evaluater.get_val_cache()
evaluater.set_records(record_actor_new)
evaluater.evaluate()
val_actor_new = evaluater.get_val_cache()
# 将记录插入数据库
mongo_helper.insert_new_reocrds(record_actor_new)
mongo_helper.insert_new_reocrds(record_actor_exist)
# 将今日用户新增的val更新到数据库
mongo_helper.update_val(val_actor_new)
mongo_helper.update_val(val_actor_exist)
record_cleaner.free_mem()
del record_cleaner
del record_grouper
del record_normalizer
del mongo_helper
del counter
del evaluater
# 生成CSV文件
util.grcount2csv()
end = time.time()
log.log('total: %s s' % (end - start))
bad_iteration += 1
continue
availability_dict = dict()
for data_index in range(data_count):
data_name = "data" + str(data_index)
availability_dict[data_name] = desired_availability
# print availability_dict
match_job_dict = scheduler.schedule(time_point, lease_period,
availability_dict)
if not match_job_dict:
print "Error : match_job_dict is none"
bad_iteration += 1
continue
# for job in match_job_dict:
# print job, match_job_dict[job]
evaluater = Evaluater(interval_tree, job_dict)
data_iterated = data_iterated + data_count
data_loss += evaluater.evaluate(time_point + lease_period,
match_job_dict)
iteration += 1
print "data_loss : ", data_loss
print "data_iterated : ", data_iterated
print "iteration, bad_iteration : ", iteration, bad_iteration
loss_rate_list.append(float(data_loss) / float(data_iterated))
print loss_rate_list
print loss_rate_list
resource_dict = {'SU-OG-CE': 'suogce', 'GLOW': 'glow', 'MWT2': 'mwt2'}
avail_dict = {'0.99': '099', '0.90': '090', '0.80': '080'}
file_name = resource_dict[sys.argv[1]] + '_avail_' + avail_dict[sys.argv[
2]] + '_replication_' + sys.argv[3] + '_lease_' + sys.argv[4] + '.txt'
