-
Notifications
You must be signed in to change notification settings - Fork 0
/
DTSH.py
150 lines (116 loc) · 4.99 KB
/
DTSH.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
from utils.tools import *
from network import *
import os
import torch
import torch.optim as optim
import time
import numpy as np
torch.multiprocessing.set_sharing_strategy('file_system')
# DTSH(ACCV2016)
# paper [Deep Supervised Hashing with Triplet Labels](https://arxiv.org/abs/1612.03900)
# code [DTSH](https://github.com/Minione/DTSH)
def get_config():
config = {
"alpha": 5,
"lambda": 1,
# "optimizer":{"type": optim.SGD, "optim_params": {"lr": 0.05, "weight_decay": 10 ** -5}},
"optimizer": {"type": optim.RMSprop, "optim_params": {"lr": 1e-5, "weight_decay": 10 ** -5}},
"info": "[DTSH]",
"resize_size": 256,
"crop_size": 224,
"batch_size": 128,
"net": AlexNet,
# "net":ResNet,
# "dataset": "cifar10",
"dataset": "cifar10-1",
# "dataset": "cifar10-2",
# "dataset": "coco",
# "dataset": "mirflickr",
# "dataset": "voc2012",
# "dataset": "imagenet",
# "dataset": "nuswide_21",
# "dataset": "nuswide_21_m",
# "dataset": "nuswide_81_m",
"epoch": 150,
"test_map": 15,
"save_path": "save/DTSH",
# "device":torch.device("cpu"),
"device": torch.device("cuda:1"),
"bit_list": [48, 32, 24, 12],
}
config = config_dataset(config)
return config
class DTSHLoss(torch.nn.Module):
def __init__(self, config, bit):
super(DTSHLoss, self).__init__()
def forward(self, u, y, ind, config):
inner_product = u @ u.t()
s = y @ y.t() > 0
count = 0
loss1 = 0
for row in range(s.shape[0]):
# if has positive pairs and negative pairs
if s[row].sum() != 0 and (~s[row]).sum() != 0:
count += 1
theta_positive = inner_product[row][s[row] == 1]
theta_negative = inner_product[row][s[row] == 0]
triple = (theta_positive.unsqueeze(1) - theta_negative.unsqueeze(0) - config["alpha"]).clamp(min=-100,
max=50)
loss1 += -(triple - torch.log(1 + torch.exp(triple))).mean()
if count != 0:
loss1 = loss1 / count
else:
loss1 = 0
loss2 = config["lambda"] * (u - u.sign()).pow(2).mean()
return loss1 + loss2
def train_val(config, bit):
device = config["device"]
train_loader, test_loader, dataset_loader, num_train, num_test, num_dataset = get_data(config)
config["num_train"] = num_train
net = config["net"](bit).to(device)
optimizer = config["optimizer"]["type"](net.parameters(), **(config["optimizer"]["optim_params"]))
criterion = DTSHLoss(config, bit)
Best_mAP = 0
for epoch in range(config["epoch"]):
current_time = time.strftime('%H:%M:%S', time.localtime(time.time()))
print("%s[%2d/%2d][%s] bit:%d, dataset:%s, training...." % (
config["info"], epoch + 1, config["epoch"], current_time, bit, config["dataset"]), end="")
net.train()
train_loss = 0
for image, label, ind in train_loader:
image = image.to(device)
label = label.to(device)
optimizer.zero_grad()
u = net(image)
loss = criterion(u, label.float(), ind, config)
train_loss += loss.item()
loss.backward()
optimizer.step()
train_loss = train_loss / len(train_loader)
print("\b\b\b\b\b\b\b loss:%.3f" % (train_loss))
if (epoch + 1) % config["test_map"] == 0:
# print("calculating test binary code......")
tst_binary, tst_label = compute_result(test_loader, net, device=device)
# print("calculating dataset binary code.......")\
trn_binary, trn_label = compute_result(dataset_loader, net, device=device)
# print("calculating map.......")
mAP = CalcTopMap(trn_binary.numpy(), tst_binary.numpy(), trn_label.numpy(), tst_label.numpy(),
config["topK"])
if mAP > Best_mAP:
Best_mAP = mAP
if "save_path" in config:
if not os.path.exists(config["save_path"]):
os.makedirs(config["save_path"])
print("save in ", config["save_path"])
np.save(os.path.join(config["save_path"], config["dataset"] + str(mAP) + "-" + "trn_binary.npy"),
trn_binary.numpy())
torch.save(net.state_dict(),
os.path.join(config["save_path"], config["dataset"] + "-" + str(mAP) + "-model.pt"))
print("%s epoch:%d, bit:%d, dataset:%s, MAP:%.3f, Best MAP: %.3f" % (
config["info"], epoch + 1, bit, config["dataset"], mAP, Best_mAP))
print(config)
if __name__ == "__main__":
config = get_config()
print(config)
for bit in config["bit_list"]:
train_val(config, bit)