def main():
#global args, best_prec1
#args = parser.parse_args()
# create model
if args.pretrained:
print("=> using pre-trained model '{}'".format(args.arch))
model = models.__dict__[args.arch](pretrained=True)
else:
print("=> creating model '{}'".format(args.arch))
model = models.__dict__[args.arch]()
if args.arch.startswith('alexnet') or args.arch.startswith('vgg'):
model.features = torch.nn.DataParallel(model.features)
model.cuda()
else:
model = torch.nn.DataParallel(model).cuda()
# define loss function (criterion) and optimizer
criterion = nn.CrossEntropyLoss().cuda()
optimizer = torch.optim.SGD(model.parameters(),
args.lr,
momentum=0.5,
weight_decay=args.weight_decay)
# optionally resume from a checkpoint
if args.resume:
if os.path.isfile(args.resume):
print("=> loading checkpoint '{}'".format(args.resume))
checkpoint = torch.load(args.resume)
args.start_epoch = checkpoint['epoch']
best_prec1 = checkpoint['best_prec1']
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
print("=> loaded checkpoint '{}' (epoch {})".format(
args.resume, checkpoint['epoch']))
else:
print("=> no checkpoint found at '{}'".format(args.resume))
cudnn.benchmark = True
# Data loading code
traindir = os.path.join(args.data, 'train')
if not os.path.exists(traindir):
os.mkdir(traindir)
valdir = os.path.join(args.data, 'val')
if not os.path.exists(valdir):
os.mkdir(valdir)
kwargs = {'num_workers': 1, 'pin_memory': True} if args.cuda else {}
train_loader = torch.utils.data.DataLoader(
datasets.MYDATA(
'/home/hankeji/Desktop/cifar12',
train=True,
transform=transforms.Compose([
transforms.ToTensor(),
#transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.batch_size,
shuffle=True,
**kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.MYDATA(
'/home/hankeji/Desktop/cifar12',
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
#transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.batch_size,
shuffle=True,
**kwargs)
if args.evaluate:
validate(val_loader, model, criterion)
return
best_prec1 = 0
prec1 = 0
best_prec1 = 0
prec1 = 0
c = []
c = np.random.rand(args.epochs, 2)
for epoch in range(args.start_epoch, args.epochs):
adjust_learning_rate(optimizer, epoch)
print epoch
# train for one epoch
train(train_loader, model, criterion, optimizer, epoch)
# evaluate on validation set
prec1 = validate(val_loader, model, criterion)
# remember best prec@1 and save checkpoint
best_prec1 = max(prec1, best_prec1)
if prec1 > best_prec1:
torch.save(
model,
'/home/hankeji/Desktop/papercode/tmp/DeseNet_GTSRB_40.pkl')
else:
torch.save(
model,
'/home/hankeji/Desktop/papercode/tmp/DeseNet_netbest_GTSRB_40.pkl'
)
'''
save_checkpoint({
'epoch': epoch + 1,
'arch': args.arch,
'state_dict': model.state_dict(),
'best_prec1': best_prec1,
'optimizer': optimizer.state_dict(),
}, is_best)
'''
import matplotlib.pyplot as plt
plt.plot(np.arange(args.epochs), c[:, 0], 'b', np.arange(args.epochs),
c[:, 1], 'r')
plt.show()
return 0
import numpy as np
import torch
import torch.nn as nn
import torch.utils.data
import torchvision.transforms as transforms
import torchvision.datasets as datasets
import torch.nn.functional as F
from torch.autograd import Variable
import torch.optim as optim
kwargs = {'num_workers': 1, 'pin_memory': True}
train_loader = torch.utils.data.DataLoader(
datasets.MYDATA(
'/home/hankeji/Desktop/cifar12',
train=True,
transform=transforms.Compose([
transforms.ToTensor(),
#transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=128,
shuffle=True,
**kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.MYDATA(
'/home/hankeji/Desktop/cifar12',
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
#transforms.Normalize((0.1307,), (0.3081,))
])),
])),
batch_size=1024,
shuffle=True,
**kwargs)
test_loader = torch.utils.data.DataLoader(datasets.MNIST(
'../tmp',
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
])),
batch_size=100,
shuffle=True,
**kwargs)
adv_data = torch.utils.data.DataLoader(datasets.MYDATA(
'../tmp',
train=True,
transform=transforms.Compose([
transforms.ToTensor(),
])),
batch_size=100,
shuffle=True,
**kwargs)
#pth0='/home/hankeji/Desktop/Adversarial Examples/Train-FGSM-0.2.npy'
#adv_example=torch.load('/home/hankeji/Desktop/Adversarial Examples/Cat_Train-FGSM-0.2.pkl')
class target_m(nn.Module):
def __init__(self):
super(target_m, self).__init__()
self.con1 = nn.Conv2d(1, 20, 5)
self.con2 = nn.Conv2d(20, 10, 5)
self.con3 = nn.Conv2d(10, 20, 5)