class CryptoNet_fivelayer(Module):
def __init__(self, in_dim, n_class):
super(CryptoNet_fivelayer, self).__init__()
self.conv = Conv_sec(in_dim,
5,
5,
5,
zero_padding=1,
stride=2,
method='SAME',
bit_length=bit_length)
self.sq1 = Activators_sec.Square(bit_length=bit_length)
self.fc1 = FullyConnect_sec(845, 100, bit_length=bit_length)
self.sq2 = Activators_sec.Square(bit_length=bit_length)
self.fc2 = FullyConnect_sec(100, n_class, bit_length=bit_length)
# self.logsoftmax = Logsoftmax()
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
def forward(self, x1, x2):
in_size = x1.shape[0]
# start_time = time.time()
out_c1, out_c2 = self.conv.forward(x1, x2)
out_11, out_12 = self.sq1.forward(out_c1, out_c2)
self.conv_out_shape = out_11.shape
# print('out1shape: ',self.conv_out_shape)
out_11 = out_11.reshape(in_size, -1) # 将输出拉成一行
out_12 = out_12.reshape(in_size, -1) # 将输出拉成一行
out_fc11, out_fc12 = self.fc1.forward(out_11, out_12)
out_21, out_22 = self.sq2.forward(out_fc11, out_fc12)
out_31, out_32 = self.fc2.forward(out_21, out_22)
# end_time = time.time()
# print('time consume: ', (end_time-start_time)*1000)
out_logsoftmax_1, out_logsoftmax_2 = self.logsoftmax.forward(
out_31, out_32)
return out_logsoftmax_1, out_logsoftmax_2
# out_logsoftmax = self.logsoftmax.forward(out_31+out_32)
# return out_logsoftmax
def backward(self, dy1, dy2):
dy_logsoftmax_1, dy_logsoftmax_2 = self.logsoftmax.gradient(dy1, dy2)
dy_f31, dy_f32 = self.fc2.gradient(dy_logsoftmax_1, dy_logsoftmax_2)
dy_sq21, dy_sq22 = self.sq2.gradient(dy_f31, dy_f32)
dy_f21, dy_f22 = self.fc1.gradient(dy_sq21, dy_sq22)
dy_f21 = dy_f21.reshape(self.conv_out_shape)
dy_f22 = dy_f22.reshape(self.conv_out_shape)
dy_sq11, dy_sq12 = self.sq1.gradient(dy_f21, dy_f22)
self.conv.gradient(dy_sq11, dy_sq12)
def __init__(self, in_dim, n_class):
super(Minionn_fivelayer, self).__init__()
self.conv = Conv_sec(in_dim,
5,
5,
5,
zero_padding=2,
stride=2,
method='SAME',
bit_length=bit_length)
self.relu1 = Activators_sec.ReLU(bit_length=bit_length)
self.fc1 = FullyConnect_sec(980, 100, bit_length=bit_length)
self.relu2 = Activators_sec.ReLU(bit_length=bit_length)
self.fc2 = FullyConnect_sec(100, n_class, bit_length=bit_length)
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
def __init__(self, in_dim, n_class):
super(CryptoNet_fivelayer, self).__init__()
self.conv = Conv_sec(in_dim,
5,
5,
5,
zero_padding=1,
stride=2,
method='SAME',
bit_length=bit_length)
self.sq1 = Activators_sec.Square(bit_length=bit_length)
self.fc1 = FullyConnect_sec(845, 100, bit_length=bit_length)
self.sq2 = Activators_sec.Square(bit_length=bit_length)
self.fc2 = FullyConnect_sec(100, n_class, bit_length=bit_length)
# self.logsoftmax = Logsoftmax()
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
def __init__(self, in_dim, n_class):
super(Secureml_fivelayer, self).__init__()
self.fc0 = FullyConnect_sec(784, 128, bit_length=bit_length)
self.sq1 = Activators_sec.Square(bit_length=bit_length)
self.fc1 = FullyConnect_sec(128, 128, bit_length=bit_length)
self.sq2 = Activators_sec.Square(bit_length=bit_length)
self.fc2 = FullyConnect_sec(128, n_class, bit_length=bit_length)
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
class Secureml_fivelayer(Module):
def __init__(self, in_dim, n_class):
super(Secureml_fivelayer, self).__init__()
self.fc0 = FullyConnect_sec(784, 128, bit_length=bit_length)
self.sq1 = Activators_sec.Square(bit_length=bit_length)
self.fc1 = FullyConnect_sec(128, 128, bit_length=bit_length)
self.sq2 = Activators_sec.Square(bit_length=bit_length)
self.fc2 = FullyConnect_sec(128, n_class, bit_length=bit_length)
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
# self.logsoftmax = Logsoftmax()
def forward(self, x1, x2):
in_size = x1.shape[0]
x1 = x1.reshape(in_size, -1)
x2 = x2.reshape(in_size, -1)
# start_time = time.time()
out_c1, out_c2 = self.fc0.forward(x1, x2)
out_11, out_12 = self.sq1.forward(out_c1, out_c2)
out_fc11, out_fc12 = self.fc1.forward(out_11, out_12)
out_21, out_22 = self.sq2.forward(out_fc11, out_fc12)
out_31, out_32 = self.fc2.forward(out_21, out_22)
# end_time = time.time()
# print('time consume: ', (end_time-start_time)*1000)
# out_logsoftmax = self.logsoftmax.forward(out_31+out_32)
# print('out_3: ',out_31+out_32)
out_logsoftmax_1, out_logsoftmax_2 = self.logsoftmax.forward(
out_31, out_32)
return out_logsoftmax_1, out_logsoftmax_2
# return out_logsoftmax
def backward(self, dy1, dy2):
# dy_logsoftmax = self.logsoftmax.gradient(dy)
dy_logsoftmax_1, dy_logsoftmax_2 = self.logsoftmax.gradient(dy1, dy2)
# 生成shares
# dy_logsoftmax_1 = np.random.uniform(0,1, dy_logsoftmax.shape)
# dy_logsoftmax_2 = dy_logsoftmax - dy_logsoftmax_1
dy_f31, dy_f32 = self.fc2.gradient(dy_logsoftmax_1, dy_logsoftmax_2)
dy_sq21, dy_sq22 = self.sq2.gradient(dy_f31, dy_f32)
dy_f21, dy_f22 = self.fc1.gradient(dy_sq21, dy_sq22)
dy_sq11, dy_sq12 = self.sq1.gradient(dy_f21, dy_f22)
self.fc0.gradient(dy_sq11, dy_sq12)
class Minionn_fivelayer(Module):
def __init__(self, in_dim, n_class):
super(Minionn_fivelayer, self).__init__()
self.conv = Conv_sec(in_dim,
5,
5,
5,
zero_padding=2,
stride=2,
method='SAME',
bit_length=bit_length)
self.relu1 = Activators_sec.ReLU(bit_length=bit_length)
self.fc1 = FullyConnect_sec(980, 100, bit_length=bit_length)
self.relu2 = Activators_sec.ReLU(bit_length=bit_length)
self.fc2 = FullyConnect_sec(100, n_class, bit_length=bit_length)
self.logsoftmax = Logsoftmax_sec(bit_length=bit_length)
def forward(self, x1, x2):
in_size = x1.shape[0]
# start_time_sum = time.time()
# start_conv = time.time()
out_c1, out_c2 = self.conv.forward(x1, x2)
# end_conv = time.time()
out_11, out_12, dfc_time1 = self.relu1.forward(out_c1, out_c2)
# start_fc1 = time.time()
self.conv_out_shape = out_11.shape
# print('out1shape: ',self.conv_out_shape)
out_11 = out_11.reshape(in_size, -1) # 将输出拉成一行
out_12 = out_12.reshape(in_size, -1) # 将输出拉成一行
out_fc11, out_fc12 = self.fc1.forward(out_11, out_12)
# end_fc1 = time.time()
out_21, out_22, dfc_time2 = self.relu2.forward(out_fc11, out_fc12)
# start_fc2 = time.time()
out_31, out_32 = self.fc2.forward(out_21, out_22)
# end_fc2 = time.time()
# end_time_sum = time.time()
# print('time consume: ', (end_conv-start_conv)*1000+(end_fc1-start_fc1)*1000+(end_fc1-start_fc1)*1000+self.relu1.offline_time+self.relu1.online_time+self.relu2.offline_time+self.relu2.online_time)
# print('time consume sum: ', (end_time_sum-start_time_sum)*1000)
out_logsoftmax_1, out_logsoftmax_2 = self.logsoftmax.forward(
out_31, out_32)
return out_logsoftmax_1, out_logsoftmax_2, dfc_time1 + dfc_time2
# out_logsoftmax = self.logsoftmax.forward(out_31+out_32)
# return out_logsoftmax
def backward(self, dy1, dy2):
dy_logsoftmax_1, dy_logsoftmax_2 = self.logsoftmax.gradient(dy1, dy2)
dy_f31, dy_f32 = self.fc2.gradient(dy_logsoftmax_1, dy_logsoftmax_2)
dy_relu21, dy_relu22 = self.relu2.gradient(dy_f31, dy_f32)
dy_f21, dy_f22 = self.fc1.gradient(dy_relu21, dy_relu22)
dy_f21 = dy_f21.reshape(self.conv_out_shape)
dy_f22 = dy_f22.reshape(self.conv_out_shape)
dy_relu11, dy_relu12 = self.relu1.gradient(dy_f21, dy_f22)
self.conv.gradient(dy_relu11, dy_relu12)
def fc_test():
bit_length = 32
batch_size = 1
in_num = 580
out_num = 580
x_numpy = np.random.randn(batch_size, in_num).astype(np.float64)
w_numpy = np.random.randn(in_num, out_num).astype(np.float64)
b_numpy = np.random.randn(out_num).astype(np.float64)
## 准备秘密分享secret sharing
x_numpy_1 = np.random.randn(batch_size, in_num).astype(np.float64)
x_numpy_2 = x_numpy-x_numpy_1
w_numpy_1 = np.random.randn(in_num, out_num).astype(np.float64)
w_numpy_2 = w_numpy-w_numpy_1
b_numpy_1 = np.random.randn(out_num).astype(np.float64)
b_numpy_2 = b_numpy-b_numpy_1
fc = FullyConnect(in_num, out_num)
fc_sec = FullyConnect_sec(in_num, out_num, bit_length=bit_length)
# 设置参数
fc.set_weight(Parameter(w_numpy, requires_grad=True))
fc_sec.set_weight_1(Parameter(w_numpy_1, requires_grad=True))
fc_sec.set_weight_2(Parameter(w_numpy_2, requires_grad=True))
fc.set_bias(Parameter(b_numpy, requires_grad=True))
fc_sec.set_bias_1(Parameter(b_numpy_1, requires_grad=True))
fc_sec.set_bias_2(Parameter(b_numpy_2, requires_grad=True))
# fc_out = fc.forward(x_numpy)
# fc_out_1, fc_out_2 = fc_sec.forward(x_numpy_1, x_numpy_2)
# print('error: \n', fc_out-(fc_out_1+fc_out_2))
test_num = 10
time_avg = 0
for i in range(test_num):
start_time_sec = time.time()
fc_out_1, fc_out_2 = fc_sec.forward(x_numpy_1, x_numpy_2)
end_time_sec = time.time()
time_avg+=(end_time_sec-start_time_sec)*1000
print('time avg sec: \n', time_avg/test_num)