server.load_model_weights(weight_path)
server.update_client_train_params(client_train_params)
server.create_clients_plus()
(x_train, y_train), (x_test, y_test) = datasets.fashion_mnist.load_data()
x_train = x_train.astype("float32") / 255
x_test = x_test.astype("float32") / 255
# Make sure images have shape (28, 28, 1)
x_train = np.expand_dims(x_train, -1)
x_test = np.expand_dims(x_test, -1)
data_handler = fed_learn.DataHandler(x_train, y_train, x_test, y_test, fed_learn.CifarProcessor(), args.debug)
data_handler.assign_data_to_clients_plus(server.clients, args.data_sampling_technique)
x_test, y_test = data_handler.preprocess(data_handler.x_test, data_handler.y_test)
test_acc = 0.1
for epoch in range(args.global_epochs):
print("Global Epoch {0} is starting".format(epoch))
server.init_for_new_epoch()
selected_clients = server.select_clients()
fed_learn.print_selected_clients(selected_clients)
for client in selected_clients:
print("Client {0} is starting the training".format(client.id))
server.send_model(client)
weight_summarizer = fed_learn.FedAvg()
server = fed_learn.Server(model_fn, weight_summarizer, args.clients + 1,
args.fraction)
weight_path = args.weights_file
if weight_path is not None:
server.load_model_weights(weight_path)
server.update_client_train_params(client_train_params)
#create the n+1 th device
server.create_clients_plus()
(x_train, y_train), (x_test, y_test) = datasets.cifar10.load_data()
data_handler = fed_learn.DataHandler(x_train, y_train, x_test, y_test,
fed_learn.CifarProcessor(), args.debug)
data_handler.assign_data_to_clients_plus(server.clients,
args.data_sampling_technique, 0.01)
x_test, y_test = data_handler.preprocess(data_handler.x_test,
data_handler.y_test)
test_acc = 0.1
for epoch in range(args.global_epochs):
print("Global Epoch {0} is starting".format(epoch))
server.init_for_new_epoch()
selected_clients = server.select_clients()
fed_learn.print_selected_clients(selected_clients)
for client in selected_clients:
print("Client {0} is starting the training".format(client.id))