sampling(bigimage,
biglabel,
numm + 1,
opt.dataroot + '/data',
rand_num=56)
if opt.resume == 'Y':
# Load state dicts
netG_A2B.load_state_dict(torch.load(opt.generator_A2B))
netG_B2A.load_state_dict(torch.load(opt.generator_B2A))
netD_A.load_state_dict(torch.load(opt.discriminator_A))
netD_B.load_state_dict(torch.load(opt.discriminator_B))
# Set model's test mode
netG_A2B.eval()
netG_B2A.eval()
netD_A.eval()
netD_B.eval()
else:
netG_A2B.apply(weights_init_normal)
netG_B2A.apply(weights_init_normal)
netD_A.apply(weights_init_normal)
netD_B.apply(weights_init_normal)
# Lossess
criterion_GAN = torch.nn.MSELoss()
criterion_cycle = torch.nn.L1Loss()
criterion_identity = torch.nn.L1Loss()
criterion_BCE = torch.nn.BCEWithLogitsLoss()
# Optimizers & LR schedulers
print('Saving Features')
if not os.path.exists(feature_file):
dataloader = torch.utils.data.DataLoader(dataset, batch_size=opt.batch_size,
shuffle=True, num_workers=opt.num_workers)
netD = Discriminator(opt.ndf, opt.nc, opt.filters, opt.strides, opt.padding)
netD.cuda()
epoch = 10
netD.load_state_dict(torch.load(opt.model_path + 'netD_epoch_{}.pth'.format(epoch)))
print(netD)
netD.eval()
n_features = 4096 # 1024x2x2
save_features(dataloader, opt.batch_size, n_features, feature_file)
print('Load Features')
data = np.loadtxt(feature_file, dtype=np.float16)
features, labels = data[:, : -1], data[:, -1: ]
shape = features.shape
print('Data has {} samples and {} features '.format(shape[0], shape[1]))
X_train, X_test, y_train, y_test = train_test_split(features, labels, test_size=0.33, random_state=42)
print('Train SVM')
clf = svm.SVC(decision_function_shape='ovo')