modelu.empty_all()
modelz.load_state_dict(model.state_dict())
# if not args.raw_train:
# modelz.load_state_dict()
# model_feature = torch.nn.DataParallel(model_feature, device_ids=range(args.ngpu))
if args.cuda:
model.cuda()
modelu.cuda()
modelz.cuda()
# optimizer = optim.Adam(model.parameters(), lr=args.lr)
optimizer = optim.SGD(model.parameters(), lr=args.lr, weight_decay=args.wd, momentum=0.9)
# optim_pred = optim.Adam(list(model_feature.parameters()) + list(model_pred.parameters()), lr=args.lr)
# optim_distortion = optim.Adam(model_distortion.parameters())
decreasing_lr = list(map(int, args.decreasing_lr.split(',')))
# crossloss = nn.CrossEntropyLoss
print('decreasing_lr: ' + str(decreasing_lr))
best_acc, old_file = 0, None
t_begin = time.time()
best_train_acc = 0.
best_dist = np.inf
f = open(os.path.join(args.loaddir, args.exp_logger), "a+")
f.write('prune_ratio: %f, quantize_bits: %d, attack: %s-%d\n' % (args.prune_ratio, args.quantize_bits, args.attack_algo, args.attack_eps))
modelu, modelz = CLdense(), CLdense()
modelu.empty_all()
modelz.load_state_dict(model.state_dict())
# if not args.raw_train:
# modelz.load_state_dict()
# model_feature = torch.nn.DataParallel(model_feature, device_ids=range(args.ngpu))
if args.cuda:
model.cuda()
modelu.cuda()
modelz.cuda()
# optimizer = optim.Adam(model.parameters(), lr=args.lr)
optimizer = optim.SGD(model.parameters(),
lr=args.lr,
weight_decay=args.wd,
momentum=0.9)
# optim_pred = optim.Adam(list(model_feature.parameters()) + list(model_pred.parameters()), lr=args.lr)
# optim_distortion = optim.Adam(model_distortion.parameters())
decreasing_lr = list(map(int, args.decreasing_lr.split(',')))
# crossloss = nn.CrossEntropyLoss
print('decreasing_lr: ' + str(decreasing_lr))
best_acc, old_file = 0, None
t_begin = time.time()
best_train_acc = 0.
best_dist = np.inf