-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
248 lines (221 loc) · 10.6 KB
/
main.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
import os
import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader
from datetime import datetime
from parser import args
from data_utils import Tourism, Preprocessing, Congestion_Dataset
from model_congestion.MF import MatrixFactorization
from model_congestion.GMF import GeneralizedMatrixFactorization
from model_visitor.GMF import GMF
from model_visitor.MLP import MLP
from model_visitor.NeuMF import NeuMF
from evaluate import RMSE, RMSE_con
import warnings
warnings.filterwarnings('ignore')
############################# CONFIGURATION #############################
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
print('device:', device)
print('Model Name:', format(args.model_name))
if args.shuffle == 1:
print('Stratify Split')
else:
print("Each Year Split")
# argparse doesn't support boolean type
use_pretrain = True if args.use_pretrain == 'True' else False
save_model = True if args.save_model == 'True' else False
# select rating
rating_name = 'visitor' if args.target == 'v' else('congestion_1' if args.target == 'c1' else 'congestion_2')
print(f'Selected target is {rating_name}')
pretrain_dir = 'pretrain_model'
if not os.path.exists(pretrain_dir):
os.mkdir(pretrain_dir)
############################## PREPARE DATASET ##########################
data = Preprocessing(shuffle=args.shuffle)
num_destination, num_time, num_sex, num_age, num_dayofweek, num_month, num_day = data.get_num()
# normalization되어 있는 train/test df 불러오기
train_df, test_df = data.preprocessing()
if rating_name == 'visitor':
train_dataset = Tourism(train_df, rating_name)
test_dataset = Tourism(test_df, rating_name)
train_dataloader = DataLoader(dataset=train_dataset, batch_size=args.batch, shuffle=False, drop_last=True)
test_dataloader = DataLoader(dataset=test_dataset, batch_size=args.batch, shuffle=False, drop_last=True)
else:
train_dataset = Congestion_Dataset(train_df, rating_name)
test_dataset = Congestion_Dataset(test_df, rating_name)
train_dataloader = DataLoader(dataset=train_dataset, batch_size=args.batch, shuffle=False, drop_last=True)
test_dataloader = DataLoader(dataset=test_dataset, batch_size=args.batch, shuffle=False, drop_last=True)
########################### CREATE MODEL #################################
# target: visitor
if rating_name == 'visitor':
if args.model_name == 'GMF':
model = GMF(num_factor=args.num_factors,
num_dayofweek=num_dayofweek,
num_time=num_time,
num_sex=num_sex,
num_age=num_age,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
use_pretrain=use_pretrain,
use_NeuMF=False,
pretrained_GMF=None)
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
elif args.model_name == 'MLP':
model = MLP(num_factor=args.num_factors,
num_layer=args.num_layers,
num_dayofweek=num_dayofweek,
num_time=num_time,
num_sex=num_sex,
num_age=num_age,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
use_pretrain=use_pretrain,
use_NeuMF=False,
pretrained_MLP=None)
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
elif args.model_name == 'NeuMF':
if use_pretrain:
GMF_dir = os.path.join(pretrain_dir, f'{args.shuffle}_False_GMF_{args.epochs}_{args.batch}_{args.num_factors}_{rating_name}.pth')
MLP_dir = os.path.join(pretrain_dir, f'{args.shuffle}_False_MLP_{args.epochs}_{args.batch}_{args.num_factors}_{rating_name}.pth')
pretrained_GMF = GMF(num_factor=args.num_factors,
num_dayofweek=num_dayofweek,
num_time=num_time,
num_sex=num_sex,
num_age=num_age,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
use_pretrain=False,
use_NeuMF=False,
pretrained_GMF=None)
pretrained_MLP = MLP(num_factor=args.num_factors,
num_layer=args.num_layers,
num_dayofweek=num_dayofweek,
num_time=num_time,
num_sex=num_sex,
num_age=num_age,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
use_pretrain=False,
use_NeuMF=False,
pretrained_MLP=None)
pretrained_GMF.load_state_dict(torch.load(GMF_dir))
pretrained_MLP.load_state_dict(torch.load(MLP_dir))
# 신경망의 모든 매개변수를 고정합니다
for param in pretrained_GMF.parameters():
param.requires_grad = False
for param in pretrained_MLP.parameters():
param.requires_grad = False
else:
pretrained_GMF = None
pretrained_MLP = None
model = NeuMF(num_factor=args.num_factors,
num_layer=args.num_layers,
num_dayofweek=num_dayofweek,
num_time=num_time,
num_sex=num_sex,
num_age=num_age,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
use_pretrain=use_pretrain,
use_NeuMF=True,
pretrained_GMF=pretrained_GMF,
pretrained_MLP=pretrained_MLP)
if not use_pretrain:
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
else:
optimizer = optim.SGD(model.parameters(), lr=args.learning_rate)
# target: congestion
else:
if args.model_name == 'MF':
model = MatrixFactorization(num_dayofweek=num_dayofweek,
num_time=num_time,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
num_factor=args.num_factors)
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
elif args.model_name == 'GMF':
model = GeneralizedMatrixFactorization(num_dayofweek=num_dayofweek,
num_time=num_time,
num_month=num_month,
num_day=num_day,
num_destination=num_destination,
num_factor=args.num_factors)
optimizer = optim.Adam(model.parameters(), lr=args.learning_rate)
model.to(device)
criterion = nn.MSELoss()
########################### TRAINING #####################################
# Training visitor
if rating_name == 'visitor':
print('--------------------Train Start---------------------')
start = datetime.now()
for epoch in range(args.epochs):
total_loss = 0
model.train()
for destination, time, sex, age, dayofweek, month, day, visitor in train_dataloader:
# itemId
destination = destination.to(device)
# user information(userId)
dayofweek,time,sex,age,month,day = dayofweek.to(device),time.to(device),sex.to(device),age.to(device),month.to(device),day.to(device)
# rating(target)
visitor = visitor.to(device)
# gradient 초기화
optimizer.zero_grad()
prediction = model(dayofweek, time, sex, age, month, day, destination)
loss = criterion(prediction, visitor)
# RMSE LOSS
loss = torch.sqrt(loss)
loss.backward()
optimizer.step()
# batch마다의 loss
total_loss += loss
# evaluation
model.eval()
rmse = RMSE(model, criterion, test_dataloader, device)
print("Epoch: {}\tTRAIN Average RMSE Loss: {}\tTEST RMSE: {}".format(epoch+1, total_loss/len(train_dataloader), rmse))
end = datetime.now()
print(f'Training Time: {end-start}')
print('-------------------Train Finished-------------------')
# Training congestion_1 & congestion_2
else:
print('--------------------Train Start---------------------')
start = datetime.now()
for epoch in range(args.epochs):
total_loss = 0
model.train()
for destination, time, dayofweek, month, day, congestion in train_dataloader:
# itemId
destination = destination.to(device)
# user information(userId)
dayofweek, time, month, day = dayofweek.to(device), time.to(device), month.to(device), day.to(device)
# rating(target)
congestion = congestion.to(device)
# gradient 초기화
optimizer.zero_grad()
prediction = model(dayofweek, time, month, day, destination)
loss = criterion(prediction, congestion)
# RMSE LOSS
loss = torch.sqrt(loss)
loss.backward()
optimizer.step()
# batch마다의 loss
total_loss += loss
# evaluation
model.eval()
rmse = RMSE_con(model, criterion, test_dataloader, device)
print("Epoch: {}\tTRAIN Average RMSE Loss: {}\tTEST RMSE: {}".format(epoch + 1,
total_loss / len(train_dataloader), rmse))
end = datetime.now()
print(f'Training Time: {end - start}')
print('-------------------Train Finished-------------------')
# visitor, congestion1/2로 학습한 모델들 저장: demo 용도 + congestion1/2 dataframe 만들 용도
if save_model:
MODEL_PATH = os.path.join(pretrain_dir,
f'{args.shuffle}_{use_pretrain}_{args.model_name}_{args.epochs}_{args.batch}_{args.num_factors}_{rating_name}.pth')
torch.save(model.state_dict(), MODEL_PATH)