path_project = os.path.abspath('..')
logger = SummaryWriter('../logs')
args = args_parser()
exp_details(args)
device = 'cuda'
train_dataset, test_dataset, user_groups = get_dataset(args)
global_model = CNNCifar(args=args)
V2 = CNNCifar(args=args)
V2.to(device)
global_model.to(device)
global_model.train()
print(global_model)
global_weights = global_model.state_dict()
train_loss, test_accuracy = [], []
val_acc_list, net_list = [], []
cv_loss, cv_acc = [], []
print_every = 2
val_loss_pre, counter = 0, 0
x = 0
my_sum = 0
for epoch in tqdm(range(100)):
x += 1
local_weights, local_losses = [], []
print(f'\n | Global Training Round : {epoch+1} |\n')
path_project = os.path.abspath('..')
logger = SummaryWriter('../logs')
args = args_parser()
exp_details(args)
device = 'cuda'
# load dataset and user groups
train_dataset, test_dataset, user_groups = get_dataset(args)
global_model = CNNCifar(args=args)
new_global_model = CNNCifar(args=args)
new_global_model.to(device)
new_global_model.train()
chafen_global_model = CNNCifar(args=args)
chafen_global_model.to(device)
chafen_global_model.train()
chafen_local_model = CNNCifar(args=args)
chafen_local_model.to(device)
chafen_local_model.train()
global_model.to(device)
global_model.train()
print(global_model)
V2 = CNNCifar(args=args)
V2.to(device)
# copy weights
global_weights = global_model.state_dict()
new_global_weights = new_global_model.state_dict()
chafen_global_weights = chafen_global_model.state_dict()