if torch.cuda.is_available():
print('CUDA enabled.')
net.cuda()
print("--- Pretrained network loaded ---")
# test(net, loader_test)
# prune the weights
masks = weight_prune(net, param['pruning_perc'])
net.set_masks(masks)
net = nn.DataParallel(net)
print("--- {}% parameters pruned ---".format(param['pruning_perc']))
test(net, loader_test)
# Retraining
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.RMSprop(net.parameters(), lr=param['learning_rate'],
weight_decay=param['weight_decay'])
train(net, criterion, optimizer, param, loader_train)
# Check accuracy and nonzeros weights in each layer
print("--- After retraining ---")
test(net, loader_test)
prune_rate(net)
# Save and load the entire model
torch.save(net.state_dict(), 'models/alexnet_pruned.pkl')
def main():
global args, best_prec1
args = parser.parse_args()
print(args)
# Set # classes
if args.data == 'UCF101':
num_classes = 101
else:
num_classes = 0
print('Specify the dataset to use ')
# 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'):
model = AlexNet(num_classes=num_classes)
model.features = torch.nn.DataParallel(model.features)
model.cuda()
else:
model = torch.nn.DataParallel(model).cuda()
# Modify last layer of the model
model_ft = models.resnet18(pretrained=True)
num_ftrs = model_ft.fc.in_features
model_ft.fc = nn.Linear(num_ftrs, 101)
model = model_ft.cuda()
model = torch.nn.DataParallel(model).cuda() # Using one GPU (device_ids = 1)
# print(model)
# Define loss function (criterion) and optimizer
criterion = nn.CrossEntropyLoss().cuda()
optimizer = torch.optim.SGD(model.parameters(), args.lr,
momentum=args.momentum,
weight_decay=args.weight_decay)
# Optionally resume from a checkpoint
if 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')
testdir = os.path.join(args.data, 'test')
normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
train_loader = torch.utils.data.DataLoader(
datasets.ImageFolder(traindir, transforms.Compose([
transforms.RandomCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
])),
batch_size = args.batch_size, shuffle=True,
num_workers=args.workers, pin_memory=True
)
val_loader = torch.utils.data.DataLoader(
datasets.ImageFolder(testdir, transforms.Compose([
transforms.Scale(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
normalize,
])),
batch_size = args.batch_size, shuffle=False,
num_workers=args.workers, pin_memory=True
)
if args.evaluate:
validate(val_loader, model, criterion)
return
for epoch in range(args.start_epoch, args.epochs):
adjust_learning_rate(optimizer, 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
is_best = prec1 > best_prec1
best_prec1 = max(prec1, best_prec1)
save_checkpoint({
'epoch': epoch + 1,
'arch': args.arch,
'state_dict': model.state_dict(),
'best_prec1': best_prec1,
'optimizer': optimizer.state_dict(),
}, is_best)
dtest = torchvision.datasets.CIFAR10(
root='./data', train=False, download=True, transform=transform_test
)
test_loader = torch.utils.data.DataLoader(
dtest, batch_size=BATCH_SIZE, shuffle=False
)
device = 'cuda' if torch.cuda.is_available() else 'cpu'
# 損失関数にクロスエントロピー誤差を用いる
criterion = nn.CrossEntropyLoss()
# モデルの定義
model = AlexNet(num_classes=num_classes)
# optimizerはSGD
optimizer = torch.optim.SGD(model.parameters(), lr=LR, momentum=0.9,
weight_decay=5e-4)
# 実際の学習部分
for epoch in range(1, NUM_EPOCHS + 1):
train_loss, train_acc = epoch_train(train_loader, model, optimizer,
criterion)
test_loss, test_acc = epoch_eval(test_loader, model, criterion)
print(f'EPOCH: [{epoch}/{NUM_EPOCHS}]')
print(f'TRAIN LOSS: {train_loss:.3f}, TRAIN ACC: {train_acc:.3f}')
print(f'TEST LOSS: {test_loss:.3f}, TEST ACC: {test_acc:.3f}')
# このように重みを保存する
parameters = model.state_dict()
torch.save(parameters, f'../weights/{epoch}.pth')
def main():
"""
This code is written for the pre-training of the AlexNet to implement the SA-Siam object tarcker.
SA-Siam has the two subnetwork, S-Net and A-Net,
and this pre-trained AlexNet is used for the feature extractor of S-Net.
I slightly changed the code from the pytorch examples
(https://github.com/pytorch/examples/tree/master/imagenet)
"""
global args, best_prec1
args = parser.parse_args()
args.distributed = args.world_size > 1
if args.distributed:
dist.init_process_group(backend=args.dist_backend,
init_method=args.dist_url,
world_size=args.world_size)
# create model
model = AlexNet()
model = torch.nn.parallel.DataParallel(model).cuda()
# model = torch.nn.parallel.DistributedDataParallel(model)
# define loss function (criterion) and optimizer
criterion = nn.CrossEntropyLoss().cuda()
optimizer = torch.optim.SGD(model.parameters(),
args.lr,
momentum=args.momentum,
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')
valdir = os.path.join(args.data, 'val')
normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
train_dataset = datasets.ImageFolder(
traindir,
transforms.Compose([
transforms.RandomResizedCrop(255),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
if args.distributed:
train_sampler = torch.utils.data.distributed.DistributedSampler(
train_dataset)
else:
train_sampler = None
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=args.batch_size,
shuffle=(train_sampler is None),
num_workers=args.workers,
pin_memory=True,
sampler=train_sampler)
val_loader = torch.utils.data.DataLoader(
datasets.ImageFolder(
valdir,
transforms.Compose([
# transforms.Resize((255, 255)),
transforms.RandomResizedCrop(255),
# transforms.CenterCrop(255),
transforms.ToTensor(),
normalize,
])),
batch_size=args.batch_size,
shuffle=False,
num_workers=args.workers,
pin_memory=True)
if args.evaluate:
validate(val_loader, model, criterion)
return
for epoch in range(args.start_epoch, args.epochs):
if args.distributed:
train_sampler.set_epoch(epoch)
adjust_learning_rate(optimizer, 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
is_best = prec1 > best_prec1
best_prec1 = max(prec1, best_prec1)
save_checkpoint(
{
'epoch': epoch + 1,
'state_dict': model.state_dict(),
'best_prec1': best_prec1,
'optimizer': optimizer.state_dict(),
}, is_best)
