def original_train(model_dir, model_filename):
"""
Original Training: train and save pre-trained paddle model
"""
# Step 1. load paddle net
[x, y, _, loss] = network.uci_network()
# Step 2. train
place = fluid.CPUPlace()
exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())
feeder = fluid.DataFeeder(place=place, feed_list=[x, y])
train_reader = paddle.batch(paddle.dataset.uci_housing.train(),
batch_size=network.BATCH_SIZE,
drop_last=True)
start_time = time.time()
for epoch_id in range(network.PADDLE_UPDATE_EPOCH):
step = 0
for data in train_reader():
avg_loss = exe.run(feed=feeder.feed(data), fetch_list=[loss.name])
if step % 50 == 0:
print('Epoch={}, Step={}, Loss={}'.format(
epoch_id, step, avg_loss[0]))
step += 1
end_time = time.time()
print('Paddle Training of Epoch={} Batch_size={}, cost time in seconds:{}'.
format(network.PADDLE_UPDATE_EPOCH, network.BATCH_SIZE,
(end_time - start_time)))
# Step 3. save model to update
mpc_du.save_trainable_model(exe=exe,
program=fluid.default_main_program(),
model_dir=model_dir,
model_filename=model_filename)
def train_infer_model(model_dir, model_filename):
"""
Original Training: train and save paddle inference model.
"""
# Step 1. load paddle net
[x, y, y_pre, loss] = network.uci_network()
# Step 2. train
place = fluid.CPUPlace()
exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())
feeder = fluid.DataFeeder(place=place, feed_list=[x, y])
train_reader = paddle.batch(
paddle.dataset.uci_housing.train(), batch_size=network.BATCH_SIZE, drop_last=True)
start_time = time.time()
for epoch_id in range(network.TRAIN_EPOCH):
step = 0
for data in train_reader():
avg_loss = exe.run(feed=feeder.feed(data), fetch_list=[loss.name])
if step % 50 == 0:
print('Epoch={}, Step={}, Loss={}'.format(epoch_id, step, avg_loss[0]))
step += 1
end_time = time.time()
print('For Prediction: Paddle Training of Epoch={} Batch_size={}, cost time in seconds:{}'
.format(network.TRAIN_EPOCH, network.BATCH_SIZE, (end_time - start_time)))
# Step 3. save inference model
fluid.io.save_inference_model(executor=exe,
main_program=fluid.default_main_program(),
dirname=model_dir,
model_filename=model_filename,
feeded_var_names=[x.name],
target_vars=[y_pre])
def encrypt_model_and_train(role, ip, server, port, model_save_dir,
model_filename):
"""
Load uci network and train MPC model.
"""
place = fluid.CPUPlace()
exe = fluid.Executor(place)
# Step 1. Initialize MPC environment and load paddle model network and initialize parameter.
pfl_mpc.init("aby3", role, ip, server, port)
[_, _, _, loss] = network.uci_network()
exe.run(fluid.default_startup_program())
# Step 2. TRANSPILE: encrypt default_main_program into MPC program
aby3.transpile()
# Step 3. MPC-TRAINING: model training based on MPC program.
mpc_data_dir = "../mpc_data/"
feature_file = mpc_data_dir + "house_feature"
feature_shape = (13, )
label_file = mpc_data_dir + "house_label"
label_shape = (1, )
if not os.path.exists('./tmp'):
os.makedirs('./tmp')
loss_file = "./tmp/uci_mpc_loss.part{}".format(role)
if os.path.exists(loss_file):
os.remove(loss_file)
batch_size = network.UCI_BATCH_SIZE
epoch_num = network.TRAIN_EPOCH
feature_name = 'x'
label_name = 'y'
loader = process_data.get_mpc_dataloader(feature_file, label_file,
feature_shape, label_shape,
feature_name, label_name, role,
batch_size)
start_time = time.time()
for epoch_id in range(epoch_num):
step = 0
for sample in loader():
mpc_loss = exe.run(feed=sample, fetch_list=[loss.name])
if step % 50 == 0:
print('Epoch={}, Step={}, Loss={}'.format(
epoch_id, step, mpc_loss))
with open(loss_file, 'ab') as f:
f.write(np.array(mpc_loss).tostring())
step += 1
end_time = time.time()
print('Mpc Training of Epoch={} Batch_size={}, cost time in seconds:{}'.
format(epoch_num, batch_size, (end_time - start_time)))
# Step 4. SAVE trained MPC model as a trainable model.
aby3.save_trainable_model(exe=exe,
model_dir=model_save_dir,
model_filename=model_filename)
print('Successfully save mpc trained model into:{}'.format(model_save_dir))