criterion = nn.CrossEntropyLoss()
model = model.cuda(device) if use_gpu else model
criterion = criterion.cuda(device) if use_gpu else criterion
# load pretrained model
if args.model2load != None:
if args.model2load.endswith('.ckpt'):
model = load_ckpt_model(model=model,
name='fc1',
ckpt_file=args.model2load,
device=device,
param2load=None)
elif args.model2load.endswith('.mat'):
model = load_mat_model(model=model,
name='fc1',
mat_file=args.model2load,
device=device,
param2load=None)
elif args.model2load.endswith('.pth'):
model = load_pth_model(model=model,
name='fc1',
pth_file=args.model2load,
device=device,
param2load=None)
else:
raise ValueError('The format of %s is not supported' %
args.model2load)
# Parse the optimizer
optimizer = parse_optim(policy=args.optim, params=model.parameters())
# Parse norms
out_dir = os.path.dirname(args.out_file)
if out_dir != '' and not os.path.exists(out_dir):
os.makedirs(out_dir)
# Parse model
model = MLP(in_dim=args.in_dim,
hidden_dims=args.hidden_dims,
out_dim=args.out_dim,
nonlinearity=args.nonlinearity)
model = model.cuda(device) if use_gpu else model
if args.model2load.endswith('.ckpt'):
ckpt = torch.load(args.model2load)
model.load_state_dict(ckpt)
elif args.model2load.endswith('.mat'):
model = load_mat_model(model=model,
name='fc1',
mat_file=args.model2load,
device=device)
elif args.model2load.endswith('.pth'):
model = load_pth_model(model=model,
name='fc1',
pth_file=args.model2load,
device=device)
else:
raise ValueError('The format of %s is not supported' % args.model2load)
# Parse norm
norm = args.norm if args.norm > 0 else np.inf
# Prepare the item to save
configs = {kwarg: value for kwarg, value in args._get_kwargs()}
tosave = {
criterion = nn.CrossEntropyLoss()
model = model.cuda(device) if use_gpu else model
criterion = criterion.cuda(device) if use_gpu else criterion
# load pretrained model
if args.model2load != None:
if args.model2load.endswith('.ckpt'):
model = load_ckpt_model(model=model,
name='lenet',
ckpt_file=args.model2load,
device=device,
param2load=args.load_mode)
elif args.model2load.endswith('.mat'):
model = load_mat_model(model=model,
name='lenet',
mat_file=args.model2load,
device=device,
param2load=args.load_mode)
elif args.model2load.endswith('.pth'):
model = load_pth_model(model=model,
name='lenet',
pth_file=args.model2load,
device=device,
param2load=args.load_mode)
else:
raise ValueError('The format of %s is not supported' %
args.model2load)
# Parse the optimizer
if args.frozen_mode == None:
optimizer = parse_optim(policy=args.optim, params=model.parameters())