-
Notifications
You must be signed in to change notification settings - Fork 0
/
run_none.py
253 lines (218 loc) · 10.6 KB
/
run_none.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
249
250
251
252
253
import torch
print(torch.cuda.device_count())
import torchvision
import torch.optim as optim
import torchvision.transforms as transforms
import torch.nn as nn
import torch.nn.functional as F
import pickle
import random
import argparse
import os
import cv2
import numpy as np
from PIL import Image
from torch.autograd import Variable
import torchvision.datasets as datasets
from models import nin,resnet,resnet_more_activation,vgg
import math
import jenkspy
from none import compromised_detection,reset_neurons,test_utils
import get_data
import test_natral_trojan
print(torch.cuda.device_count())
parser = argparse.ArgumentParser(description='PyTorch CIFAR10 and 100 Training')
parser.add_argument('--save_path_root', default='./ckpt/', type=str, metavar='PATH')
parser.add_argument('--model_filepath', default=None, type=str, metavar='PATH')
parser.add_argument('--epoch_num_1', type=int, default=320)
parser.add_argument('--epoch_num_2', type=int, default=20)
parser.add_argument('--round_num', type=int, default=1)
parser.add_argument('--lr_decay_every', type=int, default=80)
parser.add_argument('--arch', type=str, default="nin")
parser.add_argument('--base_lr', type=float, default=0.1)
parser.add_argument('--none_lr', type=float, default=1e-4)
parser.add_argument('--wd', type=float, default=1)
parser.add_argument('--poison_type', type=str, default="single_target")
parser.add_argument('--poison_rate', type=float, default=0.05)
parser.add_argument('--trigger_size', type=int, default=3)
parser.add_argument('--attack_target', type=int, default=0)
parser.add_argument('--dataset', type=str, default="cifar10")
parser.add_argument('--max_reset_fraction', type=float, default=0.05)
parser.add_argument('--lamda_l', type=float, default=0.1)
parser.add_argument('--lamda_h', type=float, default=0.9)
parser.add_argument('--num_for_detect_biased', type=float, default=-1)
parser.add_argument("--m_gpus", action='store_true')
args = parser.parse_args()
if args.model_filepath:
model_pt_name = args.model_filepath.split("/")[-1]
model_dir_name = args.model_filepath.split(model_pt_name)[0]
print("finetune")
else:
print("train from scratch")
if args.wd:
SAVE_PATH = args.save_path_root+args.dataset+"/"+args.arch+"/"+args.poison_type+"_pRate" \
+str(args.poison_rate)+"_tSize"+str(args.trigger_size)+"_aTarget"+str(args.attack_target)+"_wd" \
+"/"+"lr"+str(args.none_lr)+"_epochNum1"+str(args.epoch_num_1)+"_epochNum2"+str(args.epoch_num_2)+"_decayEvery"+str(args.lr_decay_every)+"/"
else:
SAVE_PATH = args.save_path_root+args.dataset+"/"+args.arch+"/"+args.poison_type+"_pRate" \
+str(args.poison_rate)+"_tSize"+str(args.trigger_size)+"_aTarget"+str(args.attack_target) \
+"/"+"lr"+str(args.none_lr)+"_epochNum1"+str(args.epoch_num_1)+"_epochNum2"+str(args.epoch_num_2)+"_decayEvery"+str(args.lr_decay_every)+"/"
SAVE_PATH = SAVE_PATH + "_MaxPerLayerF"+str(args.max_reset_fraction)+ "_lamda_l"+str(args.lamda_l)+"_lamda_h"+str(args.lamda_h)+"num_for_detect_biased"+str(args.num_for_detect_biased)+"/"
if not os.path.exists( SAVE_PATH):
os.makedirs(SAVE_PATH)
logger_path = SAVE_PATH + "log.txt"
logger_file = open(logger_path, 'w')
base_lr = args.base_lr
criterion = nn.CrossEntropyLoss()
num_classes,trainset,trainloader,testloader,testloader_poisoned,trainloader_no_shuffle = get_data.get_dataloaders(args.dataset,args.poison_type,args.trigger_size,args.poison_rate)
if args.arch == "nin":
print(args.dataset)
model = nin.Net(num_classes=num_classes,in_channels=3)
print(model)
for m in model.modules():
if isinstance(m, nn.Conv2d):
m.weight.data.normal_(0, 0.05)
m.bias.data.normal_(0, 0.001)
param_dict = dict(model.named_parameters())
params = []
for key, value in param_dict.items():
if key == 'classifier.20.weight':
params += [{'params':[value], 'lr':0.1 * base_lr,
'momentum':0.95, 'weight_decay':0.0001}]
elif key == 'classifier.20.bias':
params += [{'params':[value], 'lr':0.1 * base_lr,
'momentum':0.95, 'weight_decay':0.0000}]
elif 'weight' in key:
params += [{'params':[value], 'lr':1.0 * base_lr,
'momentum':0.95, 'weight_decay':0.0001}]
else:
params += [{'params':[value], 'lr':2.0 * base_lr,
'momentum':0.95, 'weight_decay':0.0000}]
optimizer = optim.SGD(params, lr=base_lr, momentum=0.9)
elif args.arch == "resnet18":
model = resnet.resnet18(num_classes=num_classes)
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "resnet18_leakyrelu":
model = resnet_more_activation.resnet18(num_classes=num_classes,activation_function="leakyrelu")
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "resnet18_elu":
model = resnet_more_activation.resnet18(num_classes=num_classes,activation_function="elu")
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "resnet18_prelu":
model = resnet_more_activation.resnet18(num_classes=num_classes,activation_function="prelu")
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "resnet18_tanhshrink":
model = resnet_more_activation.resnet18(num_classes=num_classes,activation_function="tanhshrink")
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "resnet18_softplus":
model = resnet_more_activation.resnet18(num_classes=num_classes,activation_function="softplus")
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "vgg11":
model = vgg.vgg11(num_classes=num_classes)
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
elif args.arch == "vgg16":
model = vgg.vgg16(num_classes=num_classes)
print(model)
optimizer = optim.SGD(model.parameters(), lr=base_lr, momentum=0.9, weight_decay=1e-4)
else:
print("arch is not supported")
if args.model_filepath:
try:
model.load_state_dict(torch.load(args.model_filepath))
except:
model = nn.DataParallel(model)
model.load_state_dict(torch.load(args.model_filepath))
model = model.module
if args.m_gpus:
model = nn.DataParallel(model)
model.cuda()
def fix_bn(m):
classname = m.__class__.__name__
if classname.find('BatchNorm') != -1:
m.eval()
def adjust_learning_rate(optimizer, epoch):
if epoch%args.lr_decay_every==0:
for param_group in optimizer.param_groups:
param_group['lr'] = param_group['lr'] * 0.1
def train_epoch(epoch, loader,freeze_bn=False):
model.train()
if freeze_bn:
model.apply(fix_bn)
for batch_idx, (data, target) in enumerate(loader):
data, target = Variable(data.cuda()), Variable(target.cuda())
optimizer.zero_grad()
output = model(data)
loss = criterion(output, target)
loss.backward()
optimizer.step()
if batch_idx % 100 == 0:
print('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}\tLR: {}'.format(
epoch, batch_idx * len(data), len(trainloader.dataset),
100. * batch_idx / len(trainloader), loss.item(),
optimizer.param_groups[0]['lr']))
logger_file.write('Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.6f}\tLR: {}\n'.format(
epoch, batch_idx * len(data), len(trainloader.dataset),
100. * batch_idx / len(trainloader), loss.item(),
optimizer.param_groups[0]['lr']))
if (epoch+1)%20 == 0:
torch.save(model.state_dict(), SAVE_PATH.split("latest.")[0] + "epoch" + str(epoch+1) +'.pth')
print("*"*80)
logger_file.write("*"*80+"\n")
test_utils.test(model,testloader,num_classes,criterion,SAVE_PATH,logger_file)
print("-"*80)
logger_file.write("-"*80+"\n")
if args.poison_type != "none":
test_utils.test_poisoned(model,testloader_poisoned,num_classes,criterion,SAVE_PATH,logger_file,poison_type=args.poison_type,attack_target=args.attack_target)
print("*"*80)
logger_file.write("*"*80+"\n")
if args.dataset not in ["imagenette","trojai"]:
image_size = 32
else:
image_size = 224
loader = trainloader
for epoch in range(1, args.epoch_num_1+1):
adjust_learning_rate(optimizer, epoch)
train_epoch(epoch,loader)
print("*"*80)
logger_file.write("*"*80+"\n")
test_utils.test(model,testloader,num_classes,criterion,SAVE_PATH,logger_file)
print("-"*80)
logger_file.write("-"*80+"\n")
if args.poison_type != "none":
test_utils.test_poisoned(model,testloader_poisoned,num_classes,criterion,SAVE_PATH,logger_file,poison_type=args.poison_type,attack_target=args.attack_target)
print("*"*80)
logger_file.write("*"*80+"\n")
#test_natral_trojan.test_natral_trojan(SAVE_PATH,image_size,args.arch,logger_file)
for param_group in optimizer.param_groups:
param_group['lr'] = args.none_lr
for round_id in range(1, args.round_num+1):
selected_neuros, poison_sample_index = compromised_detection.analyze_neuros(model,args.arch,args.max_reset_fraction, \
args.lamda_l,args.lamda_h, num_classes, \
args.num_for_detect_biased, \
trainloader_no_shuffle)
indices = list(set([a for a in range(len(trainset))]) - set(poison_sample_index))
if args.poison_type != "none":
loader = torch.utils.data.DataLoader(torch.utils.data.Subset(trainset, indices), batch_size=128, shuffle=True)
model = reset_neurons.reset(model,args.arch,selected_neuros)
for epoch in range(1, args.epoch_num_2+1):
adjust_learning_rate(optimizer, epoch)
train_epoch(epoch,loader,freeze_bn=True)
print("*"*80)
logger_file.write("*"*80+"\n")
test_utils.test(model,testloader,num_classes,criterion,SAVE_PATH,logger_file)
print("-"*80)
logger_file.write("-"*80+"\n")
if args.poison_type != "none":
test_utils.test_poisoned(model,testloader_poisoned,num_classes,criterion,SAVE_PATH,logger_file,poison_type=args.poison_type,attack_target=args.attack_target)
else:
test_natral_trojan.test_natral_trojan(SAVE_PATH,image_size,arch,logger_file)
print("*"*80)
logger_file.write("*"*80+"\n")
logger_file.flush()