def main(argv):
# set constant
cube_size = [10, 10, 3]
#
parser = argparse.ArgumentParser()
parser.add_argument('-d', '--dataset', help='dataset', default='')
parser.add_argument('-g', '--height', help='frame height', default=120)
parser.add_argument('-w', '--width', help='frame width', default=160)
parser.add_argument('-t', '--task', help='task to perform', default=-1)
parser.add_argument('-c',
'--clip',
help='clip index (zero-based)',
default=-1)
parser.add_argument('-s', '--set', help='test set', default=1)
parser.add_argument('-e', '--epoch', help='epoch destination', default=0)
parser.add_argument('-m', '--model', help='start model idx', default=0)
args = vars(parser.parse_args())
#
dataset = dataset_dict[args['dataset']]
dataset['path_train'] = '%s/Train' % dataset['path']
dataset['path_test'] = '%s/Test' % dataset['path']
#
task = int(args['task'])
h = int(args['height'])
w = int(args['width'])
clip_idx = int(args['clip'])
test_set = bool(int(args['set']))
n_epoch_destination = int(args['epoch'])
model_idx_to_start = int(args['model'])
model_test = model_idx_to_start
n_row, n_col = np.array([h, w]) // cube_size[:2]
print('Selected task = %d' % task)
print('started time: %s' % datetime.datetime.now())
#
dataset['cube_dir'] = './training_saver/%s/cube_%d_%d_%d_%d_%d' % (
dataset['name'], h, w, cube_size[0], cube_size[1], cube_size[2])
if not os.path.exists(dataset['cube_dir']):
pathlib.Path(dataset['cube_dir']).mkdir(parents=True, exist_ok=True)
'''========================================='''
''' Task 1: Resize frame resolution dataset '''
'''========================================='''
if task == 1:
load_images_and_resize(dataset,
new_size=[h, w],
train=True,
force_recalc=False,
return_entire_data=False)
load_images_and_resize(dataset,
new_size=[h, w],
train=False,
force_recalc=False,
return_entire_data=False)
'''========================================='''
''' Task 2: Split cubes in dataset and save '''
'''========================================='''
if task == 2:
split_cubes(dataset,
clip_idx,
cube_size,
training_set=not test_set,
force_recalc=False,
dist_thresh=None)
'''=========================================='''
''' Task 3: Train model and check validation '''
'''=========================================='''
if task == 3:
training_cubes, training_mapping = load_all_cubes_in_set(
dataset, h, w, cube_size, training_set=True)
train_model_naive_with_batch_norm(dataset,
training_cubes,
training_mapping[:, 2],
training_mapping[:, 3],
n_row,
n_col,
n_epoch_destination,
start_model_idx=model_idx_to_start,
batch_size=256 * 12)
'''====================================='''
''' Task 4: Test model and save outputs '''
'''====================================='''
if task == 4:
sequence_n_frame = count_sequence_n_frame(dataset, test=test_set)
test_cubes, test_mapping = split_cubes(dataset,
clip_idx,
cube_size,
training_set=not test_set)
test_model_naive_with_batch_norm(dataset,
test_cubes,
test_mapping[:, 2],
test_mapping[:, 3],
n_row,
n_col,
sequence_n_frame,
clip_idx,
model_idx=model_test,
batch_size=256 * 12,
using_test_data=test_set)
'''====================================='''
''' Task 5: Calculate scores of dataset '''
'''====================================='''
if task == 5:
calc_score_full_clips(dataset,
np.array([h, w]),
cube_size,
model_test,
train=False)
calc_score_full_clips(dataset,
np.array([h, w]),
cube_size,
model_test,
train=True)
'''========================='''
''' Task -5: Plot error map '''
'''========================='''
if task == -5:
frame_idx = np.arange(16)
print('selected set:', 'Test' if test_set else 'Train')
print('selected frames:', frame_idx)
plot_error_map(dataset,
np.array([h, w]),
cube_size,
clip_idx,
frame_idx,
model_test,
score_type_idx=3,
using_test_data=test_set)
'''===================='''
''' Task 6: Evaluation '''
'''===================='''
if task == 6:
if dataset in [Belleview, Train]:
dataset['ground_truth'] = load_ground_truth_Boat(
dataset, n_clip=dataset['n_clip_test'])
elif dataset == Avenue:
dataset['ground_truth'] = load_ground_truth_Avenue(
dataset['test_mask_path'], dataset['n_clip_test'])
sequence_n_frame = count_sequence_n_frame(dataset, test=True)
labels_select_last, labels_select_first, labels_select_mid = get_test_frame_labels(
dataset, sequence_n_frame, cube_size, is_subway=False)
#
for way in range(6):
# sequence_n_frame = None
if way != 1:
continue
op = np.std
full_assess_AUC(dataset,
np.array([h, w]),
cube_size,
model_test,
labels_select_first,
sequence_n_frame=sequence_n_frame,
plot_pr_idx=None,
selected_score_estimation_way=way,
operation=op,
save_roc_pr=True)
'''============================'''
''' Task -6: Manual evaluation '''
'''============================'''
if task == -6:
if dataset in [Belleview, Train]:
dataset['ground_truth'] = load_ground_truth_Boat(
dataset, n_clip=dataset['n_clip_test'])
elif dataset == Avenue:
dataset['ground_truth'] = load_ground_truth_Avenue(
dataset['test_mask_path'], dataset['n_clip_test'])
sequence_n_frame = count_sequence_n_frame(dataset, test=True)
labels_select_last, labels_select_first, labels_select_mid = get_test_frame_labels(
dataset, sequence_n_frame, cube_size, is_subway=False)
#
for way in range(6):
# sequence_n_frame = None
if way != 1:
continue
op = np.std
#
manual_assess_AUC(dataset,
np.array([h, w]),
cube_size,
model_test,
labels_select_mid,
plot_pr_idx=None,
selected_score_estimation_way=way,
operation=op)
#
manual_assess_AUC(dataset,
np.array([h, w]),
cube_size,
model_test,
labels_select_first,
plot_pr_idx=None,
selected_score_estimation_way=way,
operation=op)
#
manual_assess_AUC(dataset,
np.array([h, w]),
cube_size,
model_test,
labels_select_last,
plot_pr_idx=None,
selected_score_estimation_way=way,
operation=op)
'''==================================='''
''' Task 7: Multiple scale evaluation '''
'''==================================='''
if task == 7:
if dataset in [Belleview, Train]:
dataset['ground_truth'] = load_ground_truth_Boat(
dataset, n_clip=dataset['n_clip_test'])
elif dataset == Avenue:
dataset['ground_truth'] = load_ground_truth_Avenue(
dataset['test_mask_path'], dataset['n_clip_test'])
sequence_n_frame = count_sequence_n_frame(dataset, test=True)
labels_select_last, labels_select_first, labels_select_mid = get_test_frame_labels(
dataset, sequence_n_frame, cube_size)
#
for way in range(6):
# sequence_n_frame = None
if way != 1:
continue
op = np.std
#
full_assess_AUC_multiple_scale(
dataset,
[np.array([120, 160]),
np.array([30, 40]),
np.array([20, 20])],
cube_size,
model_test,
labels_select_mid,
sequence_n_frame=sequence_n_frame,
selected_score_estimation_way=way,
operation=op)
#
full_assess_AUC_multiple_scale(
dataset,
[np.array([120, 160]),
np.array([30, 40]),
np.array([20, 20])],
cube_size,
model_test,
labels_select_first,
sequence_n_frame=sequence_n_frame,
selected_score_estimation_way=way,
operation=op)
#
full_assess_AUC_multiple_scale(
dataset,
[np.array([120, 160]),
np.array([30, 40]),
np.array([20, 20])],
cube_size,
model_test,
labels_select_last,
sequence_n_frame=sequence_n_frame,
selected_score_estimation_way=way,
operation=op)
'''========================='''
''' Task 08: Write video '''
''' Task 11: Save score plot'''
'''========================='''
if task == 8 or task == 11:
frame_ranges = {'Belleview': (50, 443 + 157), 'Train': (2100, 3200)}
if dataset in [Belleview, Train]:
dataset['ground_truth'] = load_ground_truth_Boat(
dataset, n_clip=dataset['n_clip_test'])
elif dataset == Avenue:
dataset['ground_truth'] = load_ground_truth_Avenue(
dataset['test_mask_path'], dataset['n_clip_test'])
write_video_result(dataset,
np.array([h, w]),
cube_size,
clip_idx,
model_test,
train=not test_set,
operation=np.std,
frame_gt=dataset['ground_truth'][clip_idx],
show_all_score=False,
frame_range=frame_ranges[dataset['name']]
if dataset in [Belleview, Train] else None,
show_clf=dataset in [Belleview, Train],
save_plot_exam_only=(task == 11))
'''======================='''
''' Task -8: Write images '''
'''======================='''
if task == -8:
write_example(dataset,
np.array([h, w]),
cube_size,
clip_idx,
model_test,
operation=np.std,
scale_video=not True,
wrapall=True)
'''============================='''
''' Task 9: Visualize G filters '''
'''============================='''
if task == 9:
visualize_filters(dataset,
cube_size,
n_row,
n_col,
model_idx=model_test)
'''============================'''
''' Task -9: Visualize weights '''
'''============================'''
if task == -9:
get_weights(dataset,
np.array([h, w]),
cube_size,
model_test,
np.std,
save_as_image=True)
'''====================================='''
''' Task 10: Convert model to visualize '''
'''====================================='''
if task == 10:
convert_model(dataset, cube_size, n_row, n_col, model_idx=model_test)
print('finished time: %s' % datetime.datetime.now())
def main(args: argparse.Namespace):
logger = CompleteLogger(args.log, args.phase)
print(args)
if args.seed is not None:
random.seed(args.seed)
torch.manual_seed(args.seed)
cudnn.deterministic = True
warnings.warn('You have chosen to seed training. '
'This will turn on the CUDNN deterministic setting, '
'which can slow down your training considerably! '
'You may see unexpected behavior when restarting '
'from checkpoints.')
cudnn.benchmark = True
# Data loading code
normalize = T.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
train_transform = T.Compose(
[T.Resize(args.resize_size),
T.ToTensor(), normalize])
val_transform = T.Compose(
[T.Resize(args.resize_size),
T.ToTensor(), normalize])
dataset = datasets.__dict__[args.data]
train_source_dataset = dataset(root=args.root,
task=args.source,
split='train',
download=True,
transform=train_transform)
train_source_loader = DataLoader(train_source_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
drop_last=True)
train_target_dataset = dataset(root=args.root,
task=args.target,
split='train',
download=True,
transform=train_transform)
train_target_loader = DataLoader(train_target_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
drop_last=True)
val_dataset = dataset(root=args.root,
task=args.target,
split='test',
download=True,
transform=val_transform)
val_loader = DataLoader(val_dataset,
batch_size=args.batch_size,
shuffle=False,
num_workers=args.workers)
train_source_iter = ForeverDataIterator(train_source_loader)
train_target_iter = ForeverDataIterator(train_target_loader)
# create model
print("=> using pre-trained model '{}'".format(args.arch))
num_factors = train_source_dataset.num_factors
backbone = models.__dict__[args.arch](pretrained=True)
bottleneck_dim = args.bottleneck_dim
if args.normalization == 'IN':
backbone = convert_model(backbone)
bottleneck = nn.Sequential(
nn.Conv2d(backbone.out_features,
bottleneck_dim,
kernel_size=3,
stride=1,
padding=1),
nn.InstanceNorm2d(bottleneck_dim),
nn.ReLU(),
)
head = nn.Sequential(
nn.Conv2d(bottleneck_dim,
bottleneck_dim,
kernel_size=3,
stride=1,
padding=1), nn.InstanceNorm2d(bottleneck_dim), nn.ReLU(),
nn.Conv2d(bottleneck_dim,
bottleneck_dim,
kernel_size=3,
stride=1,
padding=1), nn.InstanceNorm2d(bottleneck_dim), nn.ReLU(),
nn.AdaptiveAvgPool2d(output_size=(1, 1)), nn.Flatten(),
nn.Linear(bottleneck_dim, num_factors), nn.Sigmoid())
for layer in head:
if isinstance(layer, nn.Conv2d) or isinstance(layer, nn.Linear):
nn.init.normal_(layer.weight, 0, 0.01)
nn.init.constant_(layer.bias, 0)
adv_head = nn.Sequential(
nn.Conv2d(bottleneck_dim,
bottleneck_dim,
kernel_size=3,
stride=1,
padding=1), nn.InstanceNorm2d(bottleneck_dim), nn.ReLU(),
nn.Conv2d(bottleneck_dim,
bottleneck_dim,
kernel_size=3,
stride=1,
padding=1), nn.InstanceNorm2d(bottleneck_dim), nn.ReLU(),
nn.AdaptiveAvgPool2d(output_size=(1, 1)), nn.Flatten(),
nn.Linear(bottleneck_dim, num_factors), nn.Sigmoid())
for layer in adv_head:
if isinstance(layer, nn.Conv2d) or isinstance(layer, nn.Linear):
nn.init.normal_(layer.weight, 0, 0.01)
nn.init.constant_(layer.bias, 0)
regressor = ImageRegressor(backbone,
num_factors,
bottleneck=bottleneck,
head=head,
adv_head=adv_head,
bottleneck_dim=bottleneck_dim,
width=bottleneck_dim)
else:
regressor = ImageRegressor(backbone,
num_factors,
bottleneck_dim=bottleneck_dim,
width=bottleneck_dim)
regressor = regressor.to(device)
print(regressor)
mdd = MarginDisparityDiscrepancy(args.margin).to(device)
# define optimizer and lr scheduler
optimizer = SGD(regressor.get_parameters(),
args.lr,
momentum=args.momentum,
weight_decay=args.wd,
nesterov=True)
lr_scheduler = LambdaLR(
optimizer, lambda x: args.lr *
(1. + args.lr_gamma * float(x))**(-args.lr_decay))
# resume from the best checkpoint
if args.phase != 'train':
checkpoint = torch.load(logger.get_checkpoint_path('best'),
map_location='cpu')
regressor.load_state_dict(checkpoint)
# analysis the model
if args.phase == 'analysis':
train_source_loader = DataLoader(train_source_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
drop_last=True)
train_target_loader = DataLoader(train_target_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
drop_last=True)
# extract features from both domains
feature_extractor = nn.Sequential(regressor.backbone,
regressor.bottleneck,
regressor.head[:-2]).to(device)
source_feature = collect_feature(train_source_loader,
feature_extractor, device)
target_feature = collect_feature(train_target_loader,
feature_extractor, device)
# plot t-SNE
tSNE_filename = osp.join(logger.visualize_directory, 'TSNE.pdf')
tsne.visualize(source_feature, target_feature, tSNE_filename)
print("Saving t-SNE to", tSNE_filename)
# calculate A-distance, which is a measure for distribution discrepancy
A_distance = a_distance.calculate(source_feature, target_feature,
device)
print("A-distance =", A_distance)
return
if args.phase == 'test':
mae = validate(val_loader, regressor, args,
train_source_dataset.factors, device)
print(mae)
return
# start training
best_mae = 100000.
for epoch in range(args.epochs):
# train for one epoch
print("lr", lr_scheduler.get_lr())
train(train_source_iter, train_target_iter, regressor, mdd, optimizer,
lr_scheduler, epoch, args)
# evaluate on validation set
mae = validate(val_loader, regressor, args,
train_source_dataset.factors, device)
# remember best mae and save checkpoint
torch.save(regressor.state_dict(),
logger.get_checkpoint_path('latest'))
if mae < best_mae:
shutil.copy(logger.get_checkpoint_path('latest'),
logger.get_checkpoint_path('best'))
best_mae = min(mae, best_mae)
print("mean MAE {:6.3f} best MAE {:6.3f}".format(mae, best_mae))
print("best_mae = {:6.3f}".format(best_mae))
logger.close()
default=f'./output/{get_current_time()}',
help='path for saving results')
opts = arg_parser.parse_args()
device = f'cuda:{opts.gpu_id}' if opts.gpu_id > -1 else 'cpu'
# prapare folder for results
os.makedirs(opts.output_path, exist_ok=True)
gvs_net = GVSNet(opts)
gvs_net.load_state_dict(torch.load(opts.pre_trained_gvsnet), strict=True)
gvs_net.to(device)
gvs_net.eval()
if device == 'cpu':
gvs_net = convert_model(gvs_net)
dataset = DataLoader(get_dataset(opts.dataset)(opts),
batch_size=1,
shuffle=False)
saver_results = SaveResults(opts.output_path, opts.dataset)
for itr, data in tqdm.tqdm(enumerate(dataset), total=len(dataset)):
data = {k: v.to(device) for k, v in data.items()}
# Let's get a list of camera poses
# modify get_cam_poses function if you need specific camera movement
data['t_vec'], data['r_mat'] = get_cam_poses(opts.movement_type,
b_size=opts.batch_size)
data['t_vec'] = [t.to(device) for t in data['t_vec']]
data['r_mat'] = [r.to(device) for r in data['r_mat']]
# Render the scene from the chosen camera poses
def main():
global args, best_prec1
### Calculate FLOPs & Param
model = getattr(models, args.model)(args)
print(model)
if args.data in ['cifar10', 'cifar100']:
IMAGE_SIZE = 32
else:
IMAGE_SIZE = 224
n_flops, n_params = measure_model(model, IMAGE_SIZE, IMAGE_SIZE)
print('FLOPs: %.2fM, Params: %.2fM' % (n_flops / 1e6, n_params / 1e6))
args.filename = "%s_%s_%s.txt" % \
(args.model, int(n_params), int(n_flops))
del (model)
print(args)
torch.manual_seed(args.manual_seed)
# torch.cuda.manual_seed_all(args.manual_seed)
torch.backends.cudnn.deterministic = True
torch.backends.cudnn.benchmark = False
np.random.seed(args.manual_seed)
### Create model
model = getattr(models, args.model)(args)
if args.model.startswith('alexnet') or args.model.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=args.momentum,
weight_decay=args.weight_decay,
nesterov=True)
### Optionally resume from a checkpoint
if args.resume:
checkpoint = load_checkpoint(args)
if checkpoint is not None:
args.start_epoch = checkpoint['epoch'] + 1
best_prec1 = checkpoint['best_prec1']
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
### Optionally convert from a model
if args.convert_from is not None:
args.evaluate = True
state_dict = torch.load(args.convert_from)['state_dict']
model.load_state_dict(state_dict)
model = model.cpu().module
convert_model(model, args)
model = nn.DataParallel(model).cuda()
head, tail = os.path.split(args.convert_from)
tail = "converted_" + tail
torch.save({'state_dict': model.state_dict()},
os.path.join(head, tail))
### Optionally evaluate from a model
if args.evaluate_from is not None:
args.evaluate = True
state_dict = torch.load(args.evaluate_from)['state_dict']
model.load_state_dict(state_dict)
### Data loading
dataset_dir = args.dataset_dir if args.dataset_dir is not None else '../data'
if args.data == "cifar10":
normalize = transforms.Normalize(mean=[0.4914, 0.4824, 0.4467],
std=[0.2471, 0.2435, 0.2616])
train_set = datasets.CIFAR10(dataset_dir,
train=True,
download=True,
transform=transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
val_set = datasets.CIFAR10(dataset_dir,
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
normalize,
]))
elif args.data == "cifar100":
normalize = transforms.Normalize(mean=[0.5071, 0.4867, 0.4408],
std=[0.2675, 0.2565, 0.2761])
train_set = datasets.CIFAR100(dataset_dir,
train=True,
download=True,
transform=transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
val_set = datasets.CIFAR100(dataset_dir,
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
normalize,
]))
else:
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_set = datasets.ImageFolder(
traindir,
transforms.Compose([
transforms.RandomSizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
val_set = datasets.ImageFolder(
valdir,
transforms.Compose([
transforms.Scale(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
normalize,
]))
train_loader = torch.utils.data.DataLoader(train_set,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
pin_memory=True)
val_loader = torch.utils.data.DataLoader(val_set,
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):
### Train for one epoch
tr_prec1, tr_prec5, loss, lr = \
train(train_loader, model, criterion, optimizer, epoch)
### Evaluate on validation set
val_prec1, val_prec5 = validate(val_loader, model, criterion)
### Remember best prec@1 and save checkpoint
is_best = val_prec1 < best_prec1
best_prec1 = max(val_prec1, best_prec1)
model_filename = 'checkpoint_%03d.pth.tar' % epoch
save_checkpoint(
{
'epoch': epoch,
'model': args.model,
'state_dict': model.state_dict(),
'best_prec1': best_prec1,
'optimizer': optimizer.state_dict(),
}, args, is_best, model_filename,
"%.4f %.4f %.4f %.4f %.4f %.4f\n" %
(val_prec1, val_prec5, tr_prec1, tr_prec5, loss, lr))
### Convert model and test
model = model.cpu().module
convert_model(model, args)
model = nn.DataParallel(model).cuda()
print(model)
validate(val_loader, model, criterion)
n_flops, n_params = measure_model(model, IMAGE_SIZE, IMAGE_SIZE)
print('FLOPs: %.2fM, Params: %.2fM' % (n_flops / 1e6, n_params / 1e6))
return
def main():
global args, best_prec1
### Calculate FLOPs & Param
model = getattr(models, args.model)(args)
print(model)
if args.data in ['cifar10', 'cifar100']:
IMAGE_SIZE = 32
else:
IMAGE_SIZE = 224
n_flops, n_params = measure_model(model, IMAGE_SIZE, IMAGE_SIZE)
print('FLOPs: %.2fM, Params: %.2fM' % (n_flops / 1e6, n_params / 1e6))
del (model)
args.no_save_model = True
print(args)
### Create model
model = getattr(models, args.model)(args)
if args.model.startswith('alexnet') or args.model.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=args.momentum,
weight_decay=args.weight_decay,
nesterov=True)
### Optionally resume from a checkpoint
if args.resume:
checkpoint = load_checkpoint(args)
if checkpoint is not None:
args.start_epoch = checkpoint['epoch'] + 1
best_prec1 = checkpoint['best_prec1']
model.load_state_dict(checkpoint['state_dict'])
optimizer.load_state_dict(checkpoint['optimizer'])
### Optionally convert from a model and saves it to onnx file
if args.convert_from is not None:
args.evaluate = True
state_dict = torch.load(args.convert_from)['state_dict']
model.load_state_dict(state_dict)
model = model.cpu().module
convert_model(model, args)
model = model.cuda()
# head, tail = os.path.split(args.convert_from)
# tail = "converted_" + tail
# torch.save({'state_dict': model.state_dict()}, os.path.join(head, tail))
dummy_input = torch.randn(args.batch_size,
3,
IMAGE_SIZE,
IMAGE_SIZE,
device='cuda')
torch.onnx.export(model,
dummy_input,
args.convert_from + ".onnx",
verbose=True)
return
### Optionally evaluate from a model
if args.evaluate_from is not None:
args.evaluate = True
state_dict = torch.load(args.evaluate_from)['state_dict']
model.load_state_dict(state_dict)
cudnn.benchmark = True
### Data loading
if args.data == "cifar10":
norm_mean = [0.49139968, 0.48215827, 0.44653124]
norm_std = [0.24703233, 0.24348505, 0.26158768]
norm_transform = transforms.Normalize(norm_mean, norm_std)
train_set = datasets.CIFAR10('~/Documents/CIFAR-10',
train=True,
download=True,
transform=transforms.Compose([
transforms.RandomCrop(
32,
padding=4,
padding_mode='reflect'),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
norm_transform,
]))
val_set = datasets.CIFAR10('~/Documents/CIFAR-10',
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
norm_transform,
]))
elif args.data == "cifar100":
normalize = transforms.Normalize(mean=[0.5071, 0.4867, 0.4408],
std=[0.2675, 0.2565, 0.2761])
train_set = datasets.CIFAR100('../data',
train=True,
download=True,
transform=transforms.Compose([
transforms.RandomCrop(32, padding=4),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
val_set = datasets.CIFAR100('../data',
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
normalize,
]))
else:
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_set = datasets.ImageFolder(
traindir,
transforms.Compose([
transforms.RandomSizedCrop(224),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
val_set = datasets.ImageFolder(
valdir,
transforms.Compose([
transforms.Scale(256),
transforms.CenterCrop(224),
transforms.ToTensor(),
normalize,
]))
train_loader = torch.utils.data.DataLoader(train_set,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
pin_memory=True)
val_loader = torch.utils.data.DataLoader(val_set,
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):
### Train for one epoch
tr_prec1, tr_prec5, loss, lr = \
train(train_loader, model, criterion, optimizer, epoch)
### Evaluate on validation set
val_prec1, val_prec5 = validate(val_loader, model, criterion)
### Remember best prec@1
is_best = val_prec1 < best_prec1
best_prec1 = max(val_prec1, best_prec1)
save_checkpoint(
{
'epoch': epoch,
'model': args.model,
'state_dict': model.state_dict(),
'best_prec1': best_prec1,
'optimizer': optimizer.state_dict(),
}, args, is_best, args.file_name)
### Convert model and test
model = model.cpu().module
convert_model(model, args)
model = nn.DataParallel(model).cuda()
print(model)
validate(val_loader, model, criterion)
n_flops, n_params = measure_model(model, IMAGE_SIZE, IMAGE_SIZE)
print('FLOPs: %.2fM, Params: %.2fM' % (n_flops / 1e6, n_params / 1e6))
return
