def main():
device = torch.device('cpu')
model = EfficientNet(1.0, 1.0)
weights = torch.load('efficientnet_torch.pth',
map_location=torch.device('cpu'))
model.load_state_dict(weights)
model.eval()
config = CONFIG()
optimizer = optim.SGD(model.parameters(),
lr=LEARNING_RATE,
momentum=config.momentum,
weight_decay=config.weight_decay)
frame_rate_calc = 1
freq = getTickFrequency()
image_path = []
writing = 0
phoneWithHand = 0
others = 0
sleep = 0
image_path.append("./others/")
image_path.append("./phoneWithHand/")
image_path.append("./sleep/")
image_path.append("./writing/")
for j in range(0, 4):
sumsum = 0
others = 0
phoneWithHand = 0
sleep = 0
writing = 0
for i in range(0, 100):
image = Image.open(image_path[j] + str(i) + '.jpg')
trans = torchvision.transforms.ToTensor()
image = trans(image)
data = image.unsqueeze(0)
t1 = getTickCount()
scores = model(data)
t2 = getTickCount()
time1 = (t2 - t1) / freq
frame_rate_calc = 1 / time1
sumsum += frame_rate_calc
pred = scores.data.max(1)[1]
sumsum /= 100
print('others: ', end='')
print(others)
print('phoneWithHand: ', end='')
print(phoneWithHand)
print('writing: ', end='')
print(writing)
print('sleep: ', end='')
print(sleep)
print('Average FPS: ', end='')
print(sumsum)
def main():
global best_val_score
gl = get_grid_list()
for g in gl:
best_val_score = 0.0
conf = EfficientNetConfig(depth=g[0], width=g[1], resolution=g[2], num_classes=10)
run_name = conf_to_name(conf)
run = wandb.init(project='EfficientNet_small', reinit=True)
run.name = run_name
run.save()
cur_model = EfficientNet(conf)
cur_model.cuda()
wandb.watch(cur_model)
train(cur_model, run_name)
print(run_name, best_val_score)
run.finish()
def model_train(fold: int) -> None:
# Prepare Data
df = pd.read_csv(os.path.join(config.save_dir, 'split_kfold.csv'))
df_train = df[df['kfold'] != fold].reset_index(drop=True)
df_val = df[df['kfold'] == fold].reset_index(drop=True)
df_train.drop(['kfold'], axis=1).to_csv(os.path.join(
config.save_dir, f'train-kfold-{fold}.csv'), index=False)
df_val.drop(['kfold'], axis=1).to_csv(os.path.join(
config.save_dir, f'val-kfold-{fold}.csv'), index=False)
train_dataset = MnistDataset(os.path.join(config.data_dir, 'train'), os.path.join(
config.save_dir, f'train-kfold-{fold}.csv'), transforms_train)
val_dataset = MnistDataset(
os.path.join(config.data_dir, 'train'), os.path.join(config.save_dir, f'val-kfold-{fold}.csv'), transforms_test)
model = MnistModel(EfficientNet())
checkpoint_callback = ModelCheckpoint(
monitor='val_loss',
dirpath=os.path.join(save_dir, f'{fold}'),
filename='{epoch:02d}-{val_loss:.2f}.pth',
save_top_k=5,
mode='min',
)
early_stopping = EarlyStopping(
monitor='val_loss',
mode='min',
)
if config.device == 'tpu':
train_loader = DataLoader(train_dataset, batch_size=16, num_workers=10, shuffle=True)
val_loader = DataLoader(val_dataset, batch_size=2, num_workers=10, shuffle=False)
trainer = Trainer(
tpu_cores=8,
num_sanity_val_steps=-1,
deterministic=True,
max_epochs=config.epochs,
callbacks=[checkpoint_callback, early_stopping]
)
else:
train_loader = DataLoader(train_dataset, batch_size=16, num_workers=10, shuffle=True)
val_loader = DataLoader(val_dataset, batch_size=8, num_workers=10, shuffle=False)
trainer = Trainer(
gpus=1,
num_sanity_val_steps=-1,
deterministic=True,
max_epochs=config.epochs,
callbacks=[checkpoint_callback, early_stopping]
)
trainer.fit(model, train_loader, val_loader)
def __init__(self, args):
super().__init__()
self.inp = 64
self.oup = 64
self.bifpn_repeat = 2
print(args.backbone)
self.backbone = EfficientNet.from_pretrained(args)
# self.backbone.get_list_features()
self.tail = nn.ModuleList([
ConvBlock(320, self.oup, 3, 2, 1),
ConvBlock(self.oup, self.oup, 3, 2, 1)
])
self.channel_same = self.change_channel(
self.backbone.get_list_feature()[-3:])
self.BiFPN_first = BiFPN(oup=self.oup, first=True)
self.BiFPN = nn.ModuleList()
for i in range(self.bifpn_repeat - 1):
self.BiFPN.append(BiFPN(oup=self.oup, first=False))
if __name__ == '__main__':
import argparse
# from efficientnet_pytorch import EfficientNet
parser = argparse.ArgumentParser(
description='Convert TF model to PyTorch model and save for easier future loading')
parser.add_argument('--model_name', type=str, default='efficientnet-b0',
help='efficientnet-b{N}, where N is an integer 0 <= N <= 7')
parser.add_argument('--pth_file', type=str, default='efficientnet-b0.pth',
help='input PyTorch model file name')
args = parser.parse_args()
# Build model
model = EfficientNet.from_name(args.model_name)
pretrained_weights = torch.load(args.pth_file)
# model.load_state_dict(pretrained_weights)#error,key mismatched
# print(type(pretrained_weights),dir(pretrained_weights))#<class 'collections.OrderedDict'>
# for key in pretrained_weights.keys():
# print(key)
del pretrained_weights['_fc.weight']#delete unuseful weights
del pretrained_weights['_fc.bias']
# for key in pretrained_weights.keys():
# print(key)
model.load_state_dict(pretrained_weights)
# from torchsummary import summary
# summary(model.cuda(), input_size=(3, 320, 320))
transform=transform_train)
trainloader = torch.utils.data.DataLoader(trainset, batch_size=2, shuffle=True)
testset = torchvision.datasets.CIFAR10(root='./data',
train=False,
download=True,
transform=transform_test)
testloader = torch.utils.data.DataLoader(testset, batch_size=2, shuffle=False)
classes = ('plane', 'car', 'bird', 'cat', 'deer', 'dog', 'frog', 'horse',
'ship', 'truck')
device = torch.device("cuda:0")
model = efficientnet_pytorch.EfficientNet.from_pretrained('efficientnet-b0')
BPnet = BPnetwork5()
mymodel = EfficientNet.from_pretrained('efficientnet-b0')
myBPnet = BPnetwork5q()
model.to(device)
BPnet.to(device)
mymodel.to(device)
myBPnet.to(device)
# myBPnet.fc1.weight = BPnet.fc1.weight.detach().clone()
# myBPnet.fc1.bias = BPnet.fc1.bias.detach().clone()
myBPnet.fc1.load_state_dict(BPnet.fc1.state_dict())
mycriterion = CrossEntropy()
criterion = nn.CrossEntropyLoss()
optimizer_cnn = optim.RMSprop(model.parameters(), lr=0.0000256, momentum=0.9)
myoptimizer_cnn = optim.RMSprop(mymodel.parameters(),
def main():
"""
--------------------------------------------- MAIN --------------------------------------------------------
Instantiates the model plus loss function and defines the dataloaders for several datasets including some
data augmentation.
Defines the grid for a grid search on lambda_max_divrs and initial_centroid_value_multipliers which both
have a big influence on the sparsity (and respectively accuracy) of the resulting ternary networks.
Starts grid search.
"""
# Manual seed for reproducibility
torch.manual_seed(363636)
# Global instances
global args, use_cuda, device
# Instantiating the parser
args = parser.parse_args()
# Global CUDA flag
use_cuda = args.cuda and torch.cuda.is_available()
# Defining device and device's map locationo
device = torch.device("cuda" if use_cuda else "cpu")
print('chosen device: ', device)
# Building the model
if args.model == 'cifar_micronet':
print('Building MicroNet for CIFAR-100 with depth multiplier {} and width multiplier {} ...'.format(
args.dw_multps[0] ** args.phi, args.dw_multps[1] ** args.phi))
model = micronet(args.dw_multps[0] ** args.phi, args.dw_multps[1] ** args.phi)
elif args.model == 'imagenet_micronet':
print('Building MicroNet for ImageNet with depth multiplier {} and width multiplier {} ...'.format(
args.dw_multps[0] ** args.phi, args.dw_multps[1] ** args.phi))
model = image_micronet(args.dw_multps[0] ** args.phi, args.dw_multps[1] ** args.phi)
elif args.model == 'efficientnet-b1':
print('Building EfficientNet-B1 ...')
model = EfficientNet.efficientnet_b1()
elif args.model == 'efficientnet-b2':
print('Building EfficientNet-B2 ...')
model = EfficientNet.efficientnet_b2()
elif args.model == 'efficientnet-b3':
print('Building EfficientNet-B3 ...')
model = EfficientNet.efficientnet_b3()
elif args.model == 'efficientnet-b4':
print('Building EfficientNet-B4 ...')
model = EfficientNet.efficientnet_b4()
for name, param in model.named_parameters():
print('\n', name)
# Transfers model to device (GPU/CPU).
model.to(device)
# Defining loss function and printing CUDA information (if available)
if use_cuda:
print("PyTorch version: ")
print(torch.__version__)
print("CUDA Version: ")
print(torch.version.cuda)
print("cuDNN version is: ")
print(cudnn.version())
cudnn.benchmark = True
loss_fct = nn.CrossEntropyLoss().cuda()
else:
loss_fct = nn.CrossEntropyLoss()
# Dataloaders for CIFAR, ImageNet and MNIST
if args.dataset == 'CIFAR100':
print('Loading CIFAR-100 data ...')
normalize = transforms.Normalize(mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
std=[x / 255.0 for x in [63.0, 62.1, 66.7]])
kwargs = {'num_workers': args.workers, 'pin_memory': True} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(
datasets.CIFAR100(root=args.data_path, train=True, transform=transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.RandomCrop(32, 4),
transforms.ColorJitter(brightness=0.3, contrast=0.3, saturation=0.3, hue=0.075),
transforms.ToTensor(),
normalize,
Cutout(n_holes=1, length=16),
]), download=True),
batch_size=args.batch_size, shuffle=True, **kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.CIFAR100(root=args.data_path, train=False, transform=transforms.Compose([
transforms.ToTensor(),
normalize,
])),
batch_size=args.val_batch_size, shuffle=False, **kwargs)
elif args.dataset == 'ImageNet':
print('Loading ImageNet data ...')
traindir = os.path.join(args.data_path, 'train')
valdir = os.path.join(args.data_path, '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(args.image_size),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
train_loader = torch.utils.data.DataLoader(
train_dataset, batch_size=args.batch_size, shuffle=True,
num_workers=args.workers, pin_memory=True)
if model.__class__.__name__ == 'EfficientNet' or 'efficientnet' in str(args.model):
image_size = EfficientNet.get_image_size(args.model)
else:
image_size = args.image_size
val_dataset = datasets.ImageFolder(
valdir,
transforms.Compose([
transforms.Resize(image_size, interpolation=PIL.Image.BICUBIC),
transforms.CenterCrop(image_size),
transforms.ToTensor(),
normalize,
]))
val_loader = torch.utils.data.DataLoader(
val_dataset, batch_size=args.val_batch_size, shuffle=False,
num_workers=args.workers, pin_memory=True)
elif args.dataset == 'MNIST':
kwargs = {'num_workers': args.workers, 'pin_memory': True} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(
datasets.MNIST(args.data_path, train=True, download=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.MNIST(args.data_path, train=False, transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307,), (0.3081,))
])),
batch_size=args.val_batch_size, shuffle=True, **kwargs)
elif args.dataset == 'CIFAR10':
normalize = transforms.Normalize(mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
std=[x / 255.0 for x in [63.0, 62.1, 66.7]])
kwargs = {'num_workers': args.workers, 'pin_memory': True} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(
datasets.CIFAR10(root=args.data_path, train=True, transform=transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.RandomCrop(32, 4),
transforms.ToTensor(),
normalize,
]), download=True),
batch_size=args.batch_size, shuffle=True, **kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.CIFAR10(root=args.data_path, train=False, transform=transforms.Compose([
transforms.ToTensor(),
normalize,
])),
batch_size=args.val_batch_size, shuffle=False, **kwargs)
else:
raise NotImplementedError('Undefined dataset name %s' % args.dataset)
# Gridsearch on dividers for lambda_max and initial cluster center values
for initial_c_divr in args.ini_c_divrs:
for lambda_max_divr in args.lambda_max_divrs:
print('lambda_max_divr: {}, initial_c_divr: {}'.format(lambda_max_divr, initial_c_divr))
logfile = open('./model_quantization/logfiles/logfile.txt', 'a+')
logfile.write('lambda_max_divr: {}, initial_c_divr: {}'.format(lambda_max_divr, initial_c_divr))
grid_search(train_loader, val_loader, model, loss_fct, lambda_max_divr, initial_c_divr)
return pred + 1
def show_result(pred, image):
"""
Show input image with a result of the test function
"""
fig = plt.figure()
title = 'predict: ' + str(pred)
fig1 = fig.add_subplot(1, 1, 1)
fig1.set_title(title)
fig1.axis("off")
plt.imshow(image)
fig.tight_layout()
plt.show()
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model_name = 'efficientnet-b0'
image_size = EfficientNet.get_image_size(model_name)
model = EfficientNet.from_pretrained(model_name, num_classes=15)
model = model.to(device)
model.load_state_dict(torch.load('best.pt', map_location=device))
image = Image.open('sample.png').convert(
'RGB') # directory setting. Change sample.png to path of an input image
pred = test(image)
show_result(pred, image)
def BackBone_Unet(backbone_name):
up_parm_dict = {
'resnet18': [512, 256, 128, 64, 64, 64, 64, 3],
'resnet34': [512, 256, 128, 64, 64, 64, 64, 3],
'resnet50': [2048, 1024, 512, 256, 128, 64, 64, 3],
'resnet101': [2048, 1024, 512, 256, 128, 64, 64, 3],
'resnet152': [2048, 1024, 512, 256, 128, 64, 64, 3],
'densenet121': [1024, 1024, 512, 256, 128, 64, 64, 3],
'densenet161': [2204, 2104, 752, 352, 128, 64, 64, 3],
'densenet201': [1920, 1792, 512, 256, 128, 64, 64, 3],
'densenet169': [1664, 1280, 512, 256, 128, 64, 64, 3],
'efficientnet-b0': [1280, 112, 40, 24, 16, 16, 64, 3],
'efficientnet-b1': [1280, 112, 40, 24, 16, 16, 64, 3],
'efficientnet-b2': [1280, 120, 48, 24, 16, 16, 64, 3],
'efficientnet-b3': [1280, 136, 48, 32, 24, 24, 64, 3],
'efficientnet-b4': [1280, 160, 56, 32, 24, 24, 64, 3],
'efficientnet-b5': [1280, 176, 64, 40, 24, 24, 64, 3],
'efficientnet-b6': [1280, 200, 72, 40, 32, 32, 64, 3],
'efficientnet-b7': [1280, 224, 80, 48, 32, 32, 64, 3]
}
efficient_param = {
# 'efficientnet type': (width_coef, depth_coef, resolution, dropout_rate)
'efficientnet-b0': (1.0, 1.0, 224, 0.2),
'efficientnet-b1': (1.0, 1.1, 224, 0.2),
'efficientnet-b2': (1.1, 1.2, 224, 0.3),
'efficientnet-b3': (1.2, 1.4, 224, 0.3),
'efficientnet-b4': (1.4, 1.8, 224, 0.4),
'efficientnet-b5': (1.6, 2.2, 224, 0.4),
'efficientnet-b6': (1.8, 2.6, 224, 0.5),
'efficientnet-b7': (2.0, 3.1, 224, 0.5)
}
if backbone_name[0] == 'r':
if backbone_name[-2:] == '18':
model = ResNet.ResNet18()
if backbone_name[-2:] == '34':
model = ResNet.ResNet34()
if backbone_name[-2:] == '50':
model = ResNet.ResNet50()
if backbone_name[-2:] == '01':
model = ResNet.ResNet101()
if backbone_name[-2:] == '52':
model = ResNet.ResNet152()
net = Res_Unet(model=model, up_parm=up_parm_dict[backbone_name])
elif backbone_name[0] == 'd':
if backbone_name[-2:] == '21':
model = DenseNet.DenseNet121(seon=False)
if backbone_name[-2:] == '61':
model = DenseNet.DenseNet161(seon=False)
if backbone_name[-2:] == '01':
model = DenseNet.DenseNet201(seon=False)
if backbone_name[-2:] == '69':
model = DenseNet.DenseNet169(seon=False)
net = Dense_Unet(model=model, up_parm=up_parm_dict[backbone_name])
elif backbone_name[0] == 'e':
param = efficient_param[backbone_name]
model = EfficientNet.EfficientNet(param)
net = Efficient_Unet(model=model, up_parm=up_parm_dict[backbone_name])
return net
#
if __name__ == "__main__":
from config import Config
from model import EfficientNet
opt = Config()
torch.cuda.empty_cache()
device = torch.device(opt.device)
criterion = torch.nn.CrossEntropyLoss().cuda()
model_name = opt.backbone
model_save_dir = os.path.join(opt.checkpoints_dir, model_name)
if not os.path.exists(model_save_dir): os.makedirs(model_save_dir)
logger = get_logger(os.path.join(model_save_dir, 'log.log'))
logger.info('Using: {}'.format(model_name))
logger.info('InputSize: {}'.format(opt.input_size))
logger.info('optimizer: {}'.format(opt.optimizer))
logger.info('lr_init: {}'.format(opt.lr))
logger.info('batch size: {}'.format(opt.train_batch_size))
logger.info('criterion: {}'.format(opt.loss))
logger.info('Using label smooth: {}'.format(opt.use_smooth_label))
logger.info('lr_scheduler: {}'.format(opt.lr_scheduler))
logger.info('Using the GPU: {}'.format(str(opt.gpu_id)))
model = EfficientNet.from_pretrained(model_name='efficientnet-b7',
num_classes=2)
model.to(device)
optimizer = optim.AdamW(model.parameters(), lr=3e-4, weight_decay=5e-4)
#lr_scheduler = torch.optim.lr_scheduler.StepLR(optimizer, step_size=1, gamma=0.5)
lr_scheduler = torch.optim.lr_scheduler.CosineAnnealingWarmRestarts(
optimizer, T_0=3, T_mult=2, eta_min=1e-6, last_epoch=-1)
train_model(model, criterion, optimizer, lr_scheduler=lr_scheduler)
file.write(str(label_no) + '\n')
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--weights',
type=str,
default=WEIGTH_PATH,
help='the weights file you want to test')
args = parser.parse_args()
# net = resnet50()
# num_ftrs = net.fc.in_features
# net.fc = nn.Linear(num_ftrs, 38)
net = EfficientNet.from_pretrained('efficientnet-b0', num_classes=39)
net = net.cuda()
print("__Resnet load success")
net.load_state_dict(torch.load(args.weights), True)
print(net)
net.eval()
# __ Walk the exam folder
folder_path = DATA_EXAM_PATH
path_list = os.listdir(folder_path)
# Output .dat file
print('*' * 30)
train_loader = get_loader(osp.join(args.data_path, 'images'),
osp.join(args.data_path, 'list_attr_celeba.txt'),
crop_size=178,
image_size=360,
batch_size=args.batch_size,
mode='all',
num_workers=args.num_workers)
print('*' * 30)
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print('Running on device:', device)
print('*' * 30)
print('Model Architecture')
model = EfficientNet.from_name('efficientnet-b4')
model = model.to(device)
model_params = torchutils.get_model_param_count(model)
print(model)
print('Total model parameters:', model_params)
print('*' * 30)
# criterion = nn.BCEWithLogitsLoss()
criterion = nn.MultiLabelSoftMarginLoss()
optimizer = optim.SGD(model.parameters(),
lr=args.lr,
momentum=0.9,
weight_decay=args.weight_decay)
scheduler = optim.lr_scheduler.ReduceLROnPlateau(optimizer,
patience=3,
threshold=0.005,
continue
total_examples += len(target)
data, target = Variable(data), Variable(target)
data, target = data.cuda(), target.cuda()
scores = model(data)
pred = scores.data.max(1)[1]
test_correct += pred.eq(target.data).cpu().sum()
print("Predicted {} out of {} correctly".format(test_correct,
total_examples))
return 100.0 * test_correct / (float(total_examples))
if __name__ == '__main__':
torch.cuda.device(0)
model = EfficientNet(1.0, 1.0)
config = CONFIG()
model = model.cuda()
avg_loss = list()
best_accuracy = 0.0
optimizer = optim.SGD(model.parameters(),
lr=LEARNING_RATE,
momentum=config.momentum,
weight_decay=config.weight_decay)
train_acc, val_acc = list(), list()
def main():
parser = argparse.ArgumentParser(description='Transfer images to styles.')
parser.add_argument('--dataset-dir',
type=str,
dest='data_dir',
help='path to content images directory',
default='data/img_tiff')
parser.add_argument('--chk-path',
type=str,
dest='weight_path',
help='path to model weights',
default='none')
parser.add_argument(
'--device',
type=str,
help='type of inference device',
default='cuda',
)
parser.add_argument('--batch-size', type=int, dest='batch_size', default=4)
parser.add_argument('--epoch', type=int, dest='num_epoch', default=10)
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
args = parser.parse_args()
num_epoch = args.num_epoch
source_images_path = glob.glob(os.path.join(
args.data_dir, '*')) # glob.glob("arg.data_dir/*")
elbow_xray_dataset = ElbowxrayDataset(xlsx_file='data/elbow.xlsx',
root_dir='data/img_tiff/',
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Resize(size=(512,
512))
]))
dataloader_train = torch.utils.data.DataLoader(elbow_xray_dataset,
batch_size=args.batch_size,
shuffle=True)
if args.weight_path == 'none':
model = EfficientNet.from_pretrained('efficientnet-b6',
num_classes=2).to(device)
model.train()
else:
state = torch.load(args.weight_path)
model.load_state_dict(state)
model.train()
optimizer = torch.optim.Adam(model.parameters(), lr=0.0001)
criterion = torch.nn.CrossEntropyLoss() # subject due to change
for epoch in range(num_epoch):
for i, data in enumerate(dataloader_train, 0):
images, labels = data['image'].to(device), data['label'].to(device)
optimizer.zero_grad()
outputs = model(images)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
if i % 20 == 19:
print('[Epoch %d/%d, %5d] loss: %.3f' %
(epoch + 1, num_epoch, i + 1, loss / 20))
torch.save(model.state_dict(), 'ckpt/' + str(epoch) + '.pkl')
def main():
# Manual seed for reproducibility
torch.manual_seed(363636)
# Global instances
global args, use_cuda, device
# Instantiating the parser
args = parser.parse_args()
# Global CUDA flag
use_cuda = args.cuda and torch.cuda.is_available()
# Defining device and device's map locationo
device = torch.device("cuda" if use_cuda else "cpu")
print('chosen device: ', device)
# Building the model
if args.model == 'cifar_micronet':
print(
'Building MicroNet for CIFAR with depth multiplier {} and width multiplier {} ...'
.format(args.dw_multps[0]**args.phi, args.dw_multps[1]**args.phi))
if args.dataset == 'CIFAR100':
num_classes = 100
elif args.dataset == 'CIFAR10':
num_classes = 10
model = micronet(args.dw_multps[0]**args.phi,
args.dw_multps[1]**args.phi, num_classes)
elif args.model == 'image_micronet':
print(
'Building MicroNet for ImageNet with depth multiplier {} and width multiplier {} ...'
.format(args.dw_multps[0]**args.phi, args.dw_multps[1]**args.phi))
model = image_micronet(args.dw_multps[0]**args.phi,
args.dw_multps[1]**args.phi)
elif args.model == 'efficientnet-b1':
print('Building EfficientNet-B1 ...')
model = EfficientNet.efficientnet_b1()
elif args.model == 'efficientnet-b2':
print('Building EfficientNet-B2 ...')
model = EfficientNet.efficientnet_b2()
elif args.model == 'efficientnet-b3':
print('Building EfficientNet-B3 ...')
model = EfficientNet.efficientnet_b3()
elif args.model == 'efficientnet-b4':
print('Building EfficientNet-B4 ...')
model = EfficientNet.efficientnet_b4()
elif args.model == 'lenet-5':
print(
'Building LeNet-5 with depth multiplier {} and width multiplier {} ...'
.format(args.dw_multps[0]**args.phi, args.dw_multps[1]**args.phi))
model = lenet5(d_multiplier=args.dw_multps[0]**args.phi,
w_multiplier=args.dw_multps[1]**args.phi)
for name, param in model.named_parameters():
print('\n', name)
# Transfers model to device (GPU/CPU).
model.to(device)
# Defining loss function and printing CUDA information (if available)
if use_cuda:
print("PyTorch version: ")
print(torch.__version__)
print("CUDA Version: ")
print(torch.version.cuda)
print("cuDNN version is: ")
print(cudnn.version())
cudnn.benchmark = True
loss_fct = nn.CrossEntropyLoss().cuda()
else:
loss_fct = nn.CrossEntropyLoss()
# Dataloaders for CIFAR, ImageNet and MNIST
if args.dataset == 'CIFAR100':
print('Loading CIFAR-100 data ...')
normalize = transforms.Normalize(
mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
std=[x / 255.0 for x in [63.0, 62.1, 66.7]])
kwargs = {
'num_workers': args.workers,
'pin_memory': True
} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(datasets.CIFAR100(
root=args.data_path,
train=True,
transform=transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.RandomCrop(32, 4),
transforms.ColorJitter(brightness=0.3,
contrast=0.3,
saturation=0.3,
hue=0.075),
transforms.ToTensor(),
normalize,
Cutout(n_holes=1, length=16),
]),
download=True),
batch_size=args.batch_size,
shuffle=True,
**kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.CIFAR100(root=args.data_path,
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
normalize,
])),
batch_size=args.val_batch_size,
shuffle=False,
**kwargs)
elif args.dataset == 'ImageNet':
print('Loading ImageNet data ...')
traindir = os.path.join(args.data_path, 'train')
valdir = os.path.join(args.data_path, '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(args.image_size),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
normalize,
]))
train_loader = torch.utils.data.DataLoader(train_dataset,
batch_size=args.batch_size,
shuffle=True,
num_workers=args.workers,
pin_memory=True)
if model.__class__.__name__ == 'EfficientNet' or 'efficientnet' in str(
args.model):
image_size = EfficientNet.get_image_size(args.model)
else:
image_size = args.image_size
val_dataset = datasets.ImageFolder(
valdir,
transforms.Compose([
transforms.Resize(image_size, interpolation=PIL.Image.BICUBIC),
transforms.CenterCrop(image_size),
transforms.ToTensor(),
normalize,
]))
val_loader = torch.utils.data.DataLoader(
val_dataset,
batch_size=args.val_batch_size,
shuffle=False,
num_workers=args.workers,
pin_memory=True)
elif args.dataset == 'MNIST':
kwargs = {
'num_workers': args.workers,
'pin_memory': True
} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(datasets.MNIST(
args.data_path,
train=True,
download=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.MNIST(args.data_path,
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
transforms.Normalize((0.1307, ), (0.3081, ))
])),
batch_size=args.val_batch_size,
shuffle=True,
**kwargs)
elif args.dataset == 'CIFAR10':
print('Loading CIFAR-10 data ...')
normalize = transforms.Normalize(
mean=[x / 255.0 for x in [125.3, 123.0, 113.9]],
std=[x / 255.0 for x in [63.0, 62.1, 66.7]])
kwargs = {
'num_workers': args.workers,
'pin_memory': True
} if use_cuda else {}
train_loader = torch.utils.data.DataLoader(datasets.CIFAR10(
root=args.data_path,
train=True,
transform=transforms.Compose([
transforms.RandomHorizontalFlip(),
transforms.RandomCrop(32, 4),
transforms.ColorJitter(brightness=0.3,
contrast=0.3,
saturation=0.3,
hue=0.075),
transforms.ToTensor(),
normalize,
Cutout(n_holes=1, length=16),
]),
download=True),
batch_size=args.batch_size,
shuffle=True,
**kwargs)
val_loader = torch.utils.data.DataLoader(
datasets.CIFAR10(root=args.data_path,
train=False,
transform=transforms.Compose([
transforms.ToTensor(),
normalize,
])),
batch_size=args.val_batch_size,
shuffle=False,
**kwargs)
else:
raise NotImplementedError('Undefined dataset name %s' % args.dataset)
train_w_frozen_assignment(train_loader, val_loader, model, loss_fct)