Nz,
D,
G,
args,
images=images,
id_labels=id_labels,
pose_labels=pose_labels)
else:
train_single_DRGAN(Nd, Np, Nz, D, G, args, dataset=mydataset)
else:
if args.batch_size % args.images_perID == 0:
train_multiple_DRGAN(images, id_labels, pose_labels, Nd, Np,
Nz, D, G, args)
else:
print(
"Please give valid combination of batch_size, images_perID"
)
exit()
else:
# pose_code = [] # specify arbitrary pose code for every image
#pose_code = np.random.uniform(-1,1, (images.shape[0], Np))
pose_code = np.ones((images.shape[0], Np))
pose_code_ = np.zeros((images.shape[0], Np))
pose_code[range(images.shape[0]), 1] = 0
pose_code_[range(images.shape[0]), pose_labels] = 1
features = Generate_Image(images, pose_code, pose_code_, Nz, G, G,
args)
args = parser.parse_args()
# Load the Expert Network
C = resnet50_ft(
weights_path='Pretrained/VGGFace2/resnet50_ft_dims_2048.pth')
print('VGGFace2 model built successfully')
if (args.generate):
evl_args_warning(args)
print('\nLoading model from [%s]...' % args.snapshot)
checkpoint = torch.load('{}_checkpoint.pth.tar'.format(args.snapshot))
D = FNM_model.Discriminator(args)
G = FNM_model.Generator(args)
D.load_state_dict(checkpoint['D_model'])
G.load_state_dict(checkpoint['G_model'])
Generate_Image(D, G, C, args)
else:
args.snapshot_dir = os.path.join(
args.snapshot_dir,
datetime.datetime.now().strftime('%Y-%m-%d_%H-%M-%S'))
os.makedirs(args.snapshot_dir)
args_warning(args)
if args.resume:
if os.path.isfile(args.resume):
print("=> loading checkpoint '{}'".format(args.resume))
D = FNM_model.Discriminator(args)
G = FNM_model.Generator(args)
checkpoint = torch.load(args.resume)
args.start_epoch = checkpoint['epoch']
D.load_state_dict(checkpoint['D_model'])
G.load_state_dict(checkpoint['G_model'])
exit()
else:
D = multi_model.Discriminator(Nd, Np, channel_num)
G = multi_model.Generator(Np, Nz, channel_num,
args.images_perID)
else:
print('\nLoading model from [%s]...' % args.snapshot)
try:
D = torch.load('{}_D.pt'.format(args.snapshot))
G = torch.load('{}_G.pt'.format(args.snapshot))
except:
print("Sorry, This snapshot doesn't exist.")
exit()
if not (args.generate):
if not (args.multi_DRGAN):
train_single_DRGAN(dataloader, Nd, Np, Nz, D, G, args)
else:
if args.batch_size % args.images_perID == 0:
train_multiple_DRGAN(images, id_labels, pose_labels, Nd, Np,
Nz, D, G, args)
else:
print(
"Please give valid combination of batch_size, images_perID"
)
exit()
else:
image_dir = '/home/home_data/jason/DR-GAN-11/test'
Generate_Image(image_dir, Nz, Np, G, args)