-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathmain.py
169 lines (145 loc) · 9.82 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
import argparse
import os
import numpy as np
import torch
import random
from Dataset import TrainDataset, data_load, KGDataset, AKGDataset
from KGCR import *
from torch.utils.data import DataLoader
from Train import train
from Full_rank import full_ranking
# from torch.utils.tensorboard import SummaryWriter
from prettytable import PrettyTable
###############################248###########################################
def init():
parser = argparse.ArgumentParser()
parser.add_argument('--seed', type=int, default=1, help='Seed init.')
parser.add_argument('--no-cuda', action='store_true', default=False, help='Disables CUDA training.')
parser.add_argument('--data_path', default='amazon-book', help='Dataset path')
parser.add_argument('--save_file', default='test1231', help='Filename')
parser.add_argument('--PATH_weight_load', default=None, help='Loading weight filename.')
parser.add_argument('--PATH_weight_save', default=None, help='Writing weight filename.')
parser.add_argument('--prefix', default='', help='Prefix of save_file.')
parser.add_argument('--alpha', type=float, default=1.0, help='Learning rate.')
parser.add_argument('--beta', type=float, default=1.0, help='Learning rate.')
parser.add_argument('--margin', type=float, default=1.0, help='Margin.')
parser.add_argument('--l_r', type=float, default=1e-3, help='Learning rate.')
parser.add_argument('--reg_weight', type=float, default=1e-2, help='Weight_regularization.')
parser.add_argument('--model_name', default='model', help='Model Name.')
parser.add_argument('--batch_size', type=int, default=1024, help='Batch size.')
parser.add_argument('--num_epoch', type=int, default=2000, help='Epoch number.')
parser.add_argument('--num_workers', type=int, default=4, help='Workers number.')
parser.add_argument('--dim_E', type=int, default=64, help='Embedding dimension.')
parser.add_argument('--topK', type=int, default=20, help='Workers number.')
parser.add_argument('--step', type=int, default=2000, help='Workers number.')
parser.add_argument('--has_pre_trained', default='True', help='Has Pretrained Module.')
parser.add_argument('--has_transE', default='False', help='Conduct TransE.')
args = parser.parse_args()
return args
if __name__ == '__main__':
args = init()
seed = args.seed
random.seed(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed_all(seed)
torch.backends.cudnn.enabled = False
torch.backends.cudnn.benchmark = False
torch.backends.cudnn.deterministic=True
device = torch.device("cuda:0" if torch.cuda.is_available() and not args.no_cuda else "cpu")
##########################################################################################################################################
data_path = args.data_path
save_file_name = args.save_file
alpha = args.alpha
beta = args.beta
margin = args.margin
learning_rate = args.l_r
reg_weight = args.reg_weight
batch_size = args.batch_size
num_workers = args.num_workers
num_epoch = args.num_epoch
prefix = args.prefix
model_name = args.model_name
dim_E = args.dim_E
topK = args.topK
step = args.step
has_pre_trained = True if args.has_pre_trained == 'True' else False
has_transE = True if args.has_transE == 'True' else False
writer = None#SummaryWriter()
# with open(data_path+'/result/result{0}_{1}.txt'.format(l_r, reg_weight), 'w') as save_file:
# save_file.write('---------------------------------lr: {0} \t reg_weight:{1} ---------------------------------\r\n'.format(l_r, reg_weight))
##########################################################################################################################################
print('Data loading ...')
train_data, test_data, kg_data, relation_list, u_e_list, user_item_dict, user_item_dict_train, h_r_dict, num_u, num_i, num_r, num_a, att_weight = data_load(data_path)
train_dataset = TrainDataset(train_data, user_item_dict, num_i, num_u)
train_dataloader = DataLoader(train_dataset, batch_size, shuffle=True, num_workers=num_workers)
# KG_dataset = KGDataset(kg_data, relation_list, h_r_dict, num_i+num_a, num_r)
# KG_dataloader = DataLoader(KG_dataset, batch_size, shuffle=True, num_workers=num_workers)
print('Data has been loaded.')
##########################################################################################################################################
print(model_name)
model = KGCR(num_u, num_i, num_a, num_r, train_data, kg_data, relation_list, u_e_list, att_weight, reg_weight, dim_E, alpha, beta, margin).cuda()
if has_pre_trained:
pretrained_id_embed = torch.load('./datasets/pretrain/'+data_path+'/pretrained_id_embed.pt').cuda()
model.id_embedding.data = pretrained_id_embed
pretrained_item_rep = torch.load('./datasets/pretrain/'+data_path+'/item_rep.pt').cuda()
model.item_pre.data = pretrained_item_rep
pretrained_att_rep = torch.load('./datasets/pretrain/'+data_path+'/attribute.pt').cuda()
model.attribute.data = pretrained_att_rep
print('pretrained_id_embed has loaded ...')
##########################################################################################################################################
optimizer = torch.optim.Adam([{'params': model.parameters(), 'lr': learning_rate}])
##########################################################################################################################################
max_precision = 0.0
max_recall = 0.0
max_NDCG = 0.0
num_decreases = 0
max_val_result = list()
max_test_result = list()
# pt = PrettyTable()
# pt.field_names = ["Epoch", "Loss", "precision", "recall", "ndcg"]
for epoch in range(num_epoch):
loss = train(epoch, len(train_dataset), train_dataloader, model, optimizer, batch_size, writer)
if torch.isnan(loss):
print(model.result)
with open('./datasets/'+data_path+'/result_{0}.txt'.format(save_file_name), 'a') as save_file:
save_file.write('lr:{0} \t reg_weight:{1} is Nan\r\n'.format( learning_rate, reg_weight))
break
torch.cuda.empty_cache()
if epoch % 10 == 0:
test_result = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, topK, model_name, 'Test/', writer)
# pt.add_row([epoch, loss, test_result[0], test_result[1], test_result[2]])
if test_result[1] > max_recall:
max_recall = test_result[1]
max_test_result = test_result
num_decreases = 0
else:
if num_decreases > 10:
torch.save(model.result, 'result.pt')
torch.save(model.ua_rep, 'ua_rep.pt')
torch.save(model.ia_rep, 'ia_rep.pt')
torch.save(model.u_result, 'u_result.pt')
torch.save(model.hat_u_result, 'hat_u_result.pt')
torch.save(model.hat_i_result, 'hat_i_result.pt')
torch.save(model.i_result, 'i_result.pt')
test_result1 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 1, model_name, 'Test/', writer)
test_result3 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 3, model_name, 'Test/', writer)
test_result5 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 5, model_name, 'Test/', writer)
with open('./datasets/'+data_path+'/result_{0}.txt'.format(save_file_name), 'a') as save_file:
save_file.write(str(args))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result1[0], test_result1[1], test_result1[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result3[0], test_result3[1], test_result3[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result5[0], test_result5[1], test_result5[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(max_test_result[0], max_test_result[1], max_test_result[2]))
break
else:
num_decreases += 1
test_result1 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 1, model_name, 'Test/', writer)
test_result3 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 3, model_name, 'Test/', writer)
test_result5 = full_ranking(epoch, model, test_data, user_item_dict_train, None, False, step, 5, model_name, 'Test/', writer)
with open('./datasets/'+data_path+'/result_{0}.txt'.format(save_file_name), 'a') as save_file:
save_file.write(str(args))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result1[0], test_result1[1], test_result1[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result3[0], test_result3[1], test_result3[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(test_result5[0], test_result5[1], test_result5[2]))
save_file.write('\r\n-----------Test Precition:{0:.4f} Recall:{1:.4f} NDCG:{2:.4f}-----------'.format(max_test_result[0], max_test_result[1], max_test_result[2]))