PPVC.mpc

# define settings and params
program.options.cisc = True
from ml import *
program.use_trunc_pr = False
Layer.n_threads = 16
FixConv2d.use_conv2ds = True

num_labels = 7
N = 97
h = 48
w = 48
c = 1
n = 7
tot_elems = h*w*c #2304
#sfix.set_precision(12, 31)

# define architecture of the CNN
layers = [
        FixConv2d([1,48,48,1], (5,5,1,64), (64,), [1, 44, 44, 64], (1, 1), padding='VALID', tf_weight_format='True'),
        Relu([1, 44, 44, 64]),
        FixAveragePool2d((1, 44, 44, 64), (1, 20, 20, 64), (5, 5),(2,2)),
        FixConv2d((1, 20, 20, 64), ( 3, 3, 64, 64), (64,), (1, 18, 18, 64), (1, 1), padding='VALID', tf_weight_format='True'),
        Relu([1, 18, 18, 64]),
        FixConv2d((1, 18, 18, 64), (3, 3, 64,64), (64,), (1, 16, 16, 64), (1, 1), padding='VALID', tf_weight_format='True'),
        Relu([1, 16, 16, 64]),
        FixAveragePool2d((1, 16, 16, 64), (1, 7, 7, 64), (3, 3),(2,2)),
        FixConv2d((1, 7, 7, 64), ( 3, 3, 64,128), (128,), (1, 5, 5, 128), (1, 1), padding='VALID', tf_weight_format='True'),
        Relu([1, 5, 5, 128]),
        FixConv2d((1, 5, 5, 128), ( 3, 3, 128,128), (128,), (1, 3, 3, 128), (1, 1), padding='VALID', tf_weight_format='True'),
        Relu([1, 3, 3, 128]),
        FixAveragePool2d((1, 3, 3, 128), (1, 1, 1, 128), (3, 3),(2,2)),
        Dense(1,128,1024,activation='relu'),
        Dense(1,1024,1024,activation='relu'),
        Dense(1,1024,7,activation='id'),
        ImprovedApproxSoftmax((1,7))
        ]

# define variables and read secret shares
cum_emot_vals = Array(num_labels,sfix)
cum_emot_vals.assign_all(0)

#start_timer(100) # Time for frame selection protocol including reading inputs
alice = MultiArray([N,h,w,c],sfix)
alice.input_from(0)
bob = sfix.Matrix(n,N) # input will be matrix 0 1 0 0, 0 0 0 1
bob.input_from(1)

start_timer(100)
# oblivious frame selection
alice_reform = sfix.Matrix(N,tot_elems)
@for_range(N)
def _(i):
        alice_reform[i].assign_vector(alice[i].get_vector())
selected_frames = sfix.Matrix(n,tot_elems)
#selected_frames.assign_vector(bob.direct_mul(alice_reform))
selected_frames = bob.direct_mul_to_matrix(alice_reform)
stop_timer(100)

# private frame classification for all selected frames
@for_range(n)
def _(i):
        graph = Optimizer()
        graph.layers = layers 
        #layers[0].X.input_from(1)
        start_timer(200)
        layers[0].X.assign_vector(selected_frames[i].get_vector())
        for layer in layers:
                layer.input_from(1)
        start_timer(300)
        graph.forward(1)
        stop_timer(300)
        stop_timer(200)
        '''
        @for_range(7)
        def _(f):
                print_ln("At %s logit is %s", f, layers[-2].Y[0][0][f].reveal())
                        @for_range(7)
        def _(f):
                print_ln("At %s prob is %s", f, layers[-1].Y[f].reveal())
        '''
        start_timer(400)
        # secure index wise sum in secure label aggregation
        cum_emot_vals.assign_vector(cum_emot_vals.__add__(layers[-1].Y))
        stop_timer(400)

start_timer(500)
x = argmax(cum_emot_vals.get_vector())
stop_timer(500)
# secure argmax and reveal in secure label aggregation
print_ln("%s", x.reveal())
#stop_timer(500)