Using TensorFlow backend.
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_1 (InputLayer)         (None, 224, 224, 3)       0         
_________________________________________________________________
block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      
_________________________________________________________________
block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     
_________________________________________________________________
block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         
_________________________________________________________________
block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     
_________________________________________________________________
block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    
_________________________________________________________________
block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         
_________________________________________________________________
block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    
_________________________________________________________________
block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    
_________________________________________________________________
block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    
_________________________________________________________________
block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         
_________________________________________________________________
block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   
_________________________________________________________________
block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   
_________________________________________________________________
block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   
_________________________________________________________________
block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         
_________________________________________________________________
block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   
_________________________________________________________________
block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   
_________________________________________________________________
block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   
_________________________________________________________________
block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         
_________________________________________________________________
flatten (Flatten)            (None, 25088)             0         
_________________________________________________________________
fc1 (Dense)                  (None, 4096)              102764544 
_________________________________________________________________
fc2 (Dense)                  (None, 4096)              16781312  
_________________________________________________________________
predictions (Dense)          (None, 1000)              4097000   
=================================================================
Total params: 138,357,544
Trainable params: 138,357,544
Non-trainable params: 0
_________________________________________________________________
None
computing model size...
Layer block1_conv1: inp: 0.574 MB, out: 12.250 MB, params: 0.007 MB
Layer block1_conv2: inp: 12.250 MB, out: 12.250 MB, params: 0.141 MB
Layer block2_conv1: inp: 3.062 MB, out: 6.125 MB, params: 0.282 MB
Layer block2_conv2: inp: 6.125 MB, out: 6.125 MB, params: 0.563 MB
Layer block3_conv1: inp: 1.531 MB, out: 3.062 MB, params: 1.126 MB
Layer block3_conv2: inp: 3.062 MB, out: 3.062 MB, params: 2.251 MB
Layer block3_conv3: inp: 3.062 MB, out: 3.062 MB, params: 2.251 MB
Layer block4_conv1: inp: 0.766 MB, out: 1.531 MB, params: 4.502 MB
Layer block4_conv2: inp: 1.531 MB, out: 1.531 MB, params: 9.002 MB
Layer block4_conv3: inp: 1.531 MB, out: 1.531 MB, params: 9.002 MB
Layer block5_conv1: inp: 0.383 MB, out: 0.383 MB, params: 9.002 MB
Layer block5_conv2: inp: 0.383 MB, out: 0.383 MB, params: 9.002 MB
Layer block5_conv3: inp: 0.383 MB, out: 0.383 MB, params: 9.002 MB
Layer fc1: inp: 0.096 MB, out: 0.016 MB, params: 392.016 MB
Layer fc2: inp: 0.016 MB, out: 0.016 MB, params: 64.016 MB
Layer predictions: inp: 0.016 MB, out: 0.004 MB, params: 15.629 MB
Total Params: 138357544 params, 527.79 MB
Total Inputs: 9115136 params, 34.77 MB
Total Outputs: 13556712 params, 51.71 MB
transform <keras.engine.topology.InputLayer object at 0x7fac92d9fb70> (next = <keras.layers.convolutional.Conv2D object at 0x7fac92da62b0>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac92da62b0> (next = <keras.layers.convolutional.Conv2D object at 0x7fac92da6860>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac92da6860> (next = <keras.layers.pooling.MaxPooling2D object at 0x7fac92da6ef0>)
transform <keras.layers.pooling.MaxPooling2D object at 0x7fac92da6ef0> (next = <keras.layers.convolutional.Conv2D object at 0x7fac90561080>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac90561080> (next = <keras.layers.convolutional.Conv2D object at 0x7fac90561b00>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac90561b00> (next = <keras.layers.pooling.MaxPooling2D object at 0x7fac9056cf98>)
transform <keras.layers.pooling.MaxPooling2D object at 0x7fac9056cf98> (next = <keras.layers.convolutional.Conv2D object at 0x7fac905037f0>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac905037f0> (next = <keras.layers.convolutional.Conv2D object at 0x7fac905122e8>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac905122e8> (next = <keras.layers.convolutional.Conv2D object at 0x7fac90520e48>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac90520e48> (next = <keras.layers.pooling.MaxPooling2D object at 0x7fac9052cf28>)
transform <keras.layers.pooling.MaxPooling2D object at 0x7fac9052cf28> (next = <keras.layers.convolutional.Conv2D object at 0x7fac904c28d0>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac904c28d0> (next = <keras.layers.convolutional.Conv2D object at 0x7fac904cfd30>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac904cfd30> (next = <keras.layers.convolutional.Conv2D object at 0x7fac904dd908>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac904dd908> (next = <keras.layers.pooling.MaxPooling2D object at 0x7fac904e8f28>)
transform <keras.layers.pooling.MaxPooling2D object at 0x7fac904e8f28> (next = <keras.layers.convolutional.Conv2D object at 0x7fac90482a58>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac90482a58> (next = <keras.layers.convolutional.Conv2D object at 0x7fac9048d6d8>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac9048d6d8> (next = <keras.layers.convolutional.Conv2D object at 0x7fac9049a4a8>)
transform <keras.layers.convolutional.Conv2D object at 0x7fac9049a4a8> (next = <keras.layers.pooling.MaxPooling2D object at 0x7fac904b2fd0>)
transform <keras.layers.pooling.MaxPooling2D object at 0x7fac904b2fd0> (next = <keras.layers.core.Flatten object at 0x7fac904bebe0>)
transform <keras.layers.core.Flatten object at 0x7fac904bebe0> (next = <keras.layers.core.Dense object at 0x7fac9044b860>)
transform <keras.layers.core.Dense object at 0x7fac9044b860> (next = <keras.layers.core.Dense object at 0x7fac904663c8>)
transform <keras.layers.core.Dense object at 0x7fac904663c8> (next = <keras.layers.core.Dense object at 0x7fac90466cc0>)
transform <keras.layers.core.Dense object at 0x7fac90466cc0> (next = None)
_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
input_1 (InputLayer)         (None, 224, 224, 3)       0         
_________________________________________________________________
block1_conv1 (Conv2DQ)       (None, 224, 224, 64)      3584      
_________________________________________________________________
activation_q_1 (ActivationQ) (None, 224, 224, 64)      0         
_________________________________________________________________
block1_conv2 (Conv2DQ)       (None, 224, 224, 64)      73856     
_________________________________________________________________
activation_q_2 (ActivationQ) (None, 224, 224, 64)      0         
_________________________________________________________________
block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         
_________________________________________________________________
block2_conv1 (Conv2DQ)       (None, 112, 112, 128)     147712    
_________________________________________________________________
activation_q_3 (ActivationQ) (None, 112, 112, 128)     0         
_________________________________________________________________
block2_conv2 (Conv2DQ)       (None, 112, 112, 128)     295168    
_________________________________________________________________
activation_q_4 (ActivationQ) (None, 112, 112, 128)     0         
_________________________________________________________________
block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         
_________________________________________________________________
block3_conv1 (Conv2DQ)       (None, 56, 56, 256)       590336    
_________________________________________________________________
activation_q_5 (ActivationQ) (None, 56, 56, 256)       0         
_________________________________________________________________
block3_conv2 (Conv2DQ)       (None, 56, 56, 256)       1180160   
_________________________________________________________________
activation_q_6 (ActivationQ) (None, 56, 56, 256)       0         
_________________________________________________________________
block3_conv3 (Conv2DQ)       (None, 56, 56, 256)       1180160   
_________________________________________________________________
activation_q_7 (ActivationQ) (None, 56, 56, 256)       0         
_________________________________________________________________
block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         
_________________________________________________________________
block4_conv1 (Conv2DQ)       (None, 28, 28, 512)       2360320   
_________________________________________________________________
activation_q_8 (ActivationQ) (None, 28, 28, 512)       0         
_________________________________________________________________
block4_conv2 (Conv2DQ)       (None, 28, 28, 512)       4719616   
_________________________________________________________________
activation_q_9 (ActivationQ) (None, 28, 28, 512)       0         
_________________________________________________________________
block4_conv3 (Conv2DQ)       (None, 28, 28, 512)       4719616   
_________________________________________________________________
activation_q_10 (ActivationQ (None, 28, 28, 512)       0         
_________________________________________________________________
block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         
_________________________________________________________________
block5_conv1 (Conv2DQ)       (None, 14, 14, 512)       4719616   
_________________________________________________________________
activation_q_11 (ActivationQ (None, 14, 14, 512)       0         
_________________________________________________________________
block5_conv2 (Conv2DQ)       (None, 14, 14, 512)       4719616   
_________________________________________________________________
activation_q_12 (ActivationQ (None, 14, 14, 512)       0         
_________________________________________________________________
block5_conv3 (Conv2DQ)       (None, 14, 14, 512)       4719616   
_________________________________________________________________
activation_q_13 (ActivationQ (None, 14, 14, 512)       0         
_________________________________________________________________
block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         
_________________________________________________________________
flatten (Flatten)            (None, 25088)             0         
_________________________________________________________________
fc1 (DenseQ)                 (None, 4096)              205529088 
_________________________________________________________________
activation_q_14 (ActivationQ (None, 4096)              0         
_________________________________________________________________
fc2 (DenseQ)                 (None, 4096)              33562624  
_________________________________________________________________
activation_q_15 (ActivationQ (None, 4096)              0         
_________________________________________________________________
predictions (DenseQ)         (None, 1000)              8194000   
_________________________________________________________________
activation_q_16 (ActivationQ (None, 1000)              0         
=================================================================
Total params: 276,715,088
Trainable params: 276,715,088
Non-trainable params: 0
_________________________________________________________________
None
Initializing Enclave...
Enclave id: 2
<python.slalom.quant_layers.Conv2DQ object at 0x7fac92d9fc50> (None, 224, 224, 3) (None, 224, 224, 64) (3, 3, 3, 64)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac75745ba8> (None, 224, 224, 64) (None, 224, 224, 64) (3, 3, 64, 64)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac90123320> (None, 112, 112, 64) (None, 112, 112, 128) (3, 3, 64, 128)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac900c9550> (None, 112, 112, 128) (None, 112, 112, 128) (3, 3, 128, 128)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac900db9e8> (None, 56, 56, 128) (None, 56, 56, 256) (3, 3, 128, 256)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac900f7ac8> (None, 56, 56, 256) (None, 56, 56, 256) (3, 3, 256, 256)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac9008aa20> (None, 56, 56, 256) (None, 56, 56, 256) (3, 3, 256, 256)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac9013d358> (None, 28, 28, 256) (None, 28, 28, 512) (3, 3, 256, 512)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac900a6940> (None, 28, 28, 512) (None, 28, 28, 512) (3, 3, 512, 512)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac9017ed30> (None, 28, 28, 512) (None, 28, 28, 512) (3, 3, 512, 512)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac901070b8> (None, 14, 14, 512) (None, 14, 14, 512) (3, 3, 512, 512)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac90161860> (None, 14, 14, 512) (None, 14, 14, 512) (3, 3, 512, 512)
<python.slalom.quant_layers.Conv2DQ object at 0x7fac901a7908> (None, 14, 14, 512) (None, 14, 14, 512) (3, 3, 512, 512)
sum(abs(dense_w)): 187876.9375
sum(abs(dense_w)): 58957.9375
sum(abs(dense_w)): 27298.74609375
loading model in float32
loading model...
In load model with verif mode=0, verif_preproc=0
Json parsed
loading Input layer (None, 224, 224, 3)
loading Conv2D layer
in Conv2D with out_shape = (224, 224, 64)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (224, 224, 64)
loading Activation
loading activation relu
loading MaxPool2D layer
in Pool2D with window = (2, 2), stride = (2, 2), padding = 1, out_shape = (112, 112, 64), pad = (0, 0)
loading Conv2D layer
in Conv2D with out_shape = (112, 112, 128)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (112, 112, 128)
loading Activation
loading activation relu
loading MaxPool2D layer
in Pool2D with window = (2, 2), stride = (2, 2), padding = 1, out_shape = (56, 56, 128), pad = (0, 0)
loading Conv2D layer
in Conv2D with out_shape = (56, 56, 256)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (56, 56, 256)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (56, 56, 256)
loading Activation
loading activation relu
loading MaxPool2D layer
in Pool2D with window = (2, 2), stride = (2, 2), padding = 1, out_shape = (28, 28, 256), pad = (0, 0)
loading Conv2D layer
in Conv2D with out_shape = (28, 28, 512)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (28, 28, 512)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (28, 28, 512)
loading Activation
loading activation relu
loading MaxPool2D layer
in Pool2D with window = (2, 2), stride = (2, 2), padding = 1, out_shape = (14, 14, 512), pad = (0, 0)
loading Conv2D layer
in Conv2D with out_shape = (14, 14, 512)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (14, 14, 512)
loading Activation
loading activation relu
loading Conv2D layer
in Conv2D with out_shape = (14, 14, 512)
loading Activation
loading activation relu
loading MaxPool2D layer
in Pool2D with window = (2, 2), stride = (2, 2), padding = 1, out_shape = (7, 7, 512), pad = (0, 0)
loading Dense layer
loading Activation
loading activation relu
loading Dense layer
loading Activation
loading activation relu
loading Dense layer
loading Activation
loading activation softmax
model loaded
num_classes: 1000
input images: 10562571.0
total time: 0.9138 sec
predict returned!
returning...
recorded time: 0.9163356180069968
	top1 err: 0.0
	top5 err: 0.0
	process one image per 0.9163 s (0.9163 s realtime)
input images: 10388793.0
total time: 0.8163 sec
predict returned!
returning...
recorded time: 0.8170241279876791
	top1 err: 0.0
	top5 err: 0.0
	process one image per 0.8170 s (0.8170 s realtime)
input images: 10040208.0
total time: 0.8109 sec
predict returned!
returning...
recorded time: 0.8115918399998918
	top1 err: 33.3
	top5 err: 0.0
	process one image per 0.8143 s (0.8116 s realtime)
input images: 10262993.0
total time: 0.8147 sec
predict returned!
returning...
recorded time: 0.8152472390211187
	top1 err: 25.0
	top5 err: 0.0
	process one image per 0.8146 s (0.8152 s realtime)
Destroying Enclave with id: 2
