CNN模型复现7 DenseNet
7.DenseNet
DenseNet:arXiv:1608.06993v5 [cs.CV] 28 Jun 2018;https://arxiv.org/pdf/1608.06993v5.pdf
Dense Convo-
lutional Network (DenseNet),连接每一层传递到其他每一层。对于
每一层,所有前面层的特征图用作输入,它自己的特征图用作所有后续层的输入。 DenseNets 不是从极深或极宽的架构中汲取表征能力,而是通过特征重用来利用网络的潜力,产生易于训练且参数效率高的密集模型。
DenceNet的优点:
- 输入特层和本层的输出特征堆叠,作为下一层的输入,它们减轻了梯度消失问题,加强了特征传播。
- DenseBlock中特征重用,不需要重新学习冗余特征图,并大大减少了参数的数量。DenseNet 层非常窄(例如,每层 12 个过滤器),仅将一小组特征图添加到网络的“集体知识”中,并保持其余特征图不变——最终分类器根据网络中的所有特征图。
- 每一层都可以直接得到自损失函数对原始输入信号的梯度,从而导致隐式的深度监督 ,这有助于训练更深的网络架构和准确率。
- 密集连接具有正则化效果,这减少了对具有较小训练集大小的任务的过度拟合。
- Dense connectivity: xl = concatenate(Hl([x0,x1,…,xl−1]),[x0,x1,…,xl−1])
- 在DenseBlock中采用bottlleneck结构,BN-ReLU-Conv(1× 1)-BN-ReLU-Conv(3×3) .在DenseBlock中,每层增加k=32个通道,用ConV(1x1)生成4k通道数,降低输入的通道数,提高计算效率。
- DenseNet 有关此正则化的功能。
7.1 网络结构
(1) 网络结构
- Conv + Pool
- Dense block (1x1conv,4*f --> 3x3conv,f)+ Transition block(1x1conv -->–>AvgPool(2x2,s=2))
- Global Average Pool + Fully Connected layers
(2) Dense block
bn_relu_conv(1x1,4*filters)
↓
bn_relu_conv(3x3,filters)
↓
concatenate(Input)
(2)Transition block
bn_relu_conv(1x1,theta*k)
↓
AvgPool(2x2,s=2)
(3) Model diagram
Conv(f=7,s=2)
Pool(p=3,s=2)
↓
Dense block*6
↓
Transition block
↓
Dense block*2
↓
Transition block
↓
Dense block*24
↓
Transition block
↓
Dense block*16
↓
Transition block
↓
Global Average Pool
Fully Connected layers
7.2 代码
(1) 代码流程
- 导入程序包
- 编写BN_ReLU_Conv function
- 编写dense block
- Transition block
- 建立model
(2) 代码
# 1 Import
from keras import Model
import keras
from keras.layers import Input,Conv2D,Dense,Concatenate,ReLU,BatchNormalization,AvgPool2D,MaxPool2D,GlobalAvgPool2D
# 2 BN-ReLU-Conv
def bn_relu_conv(x,filters,kernel_size):
'''
BN+ReLU+Conv
'''
x = BatchNormalization()(x)
x = ReLU()(x)
x = Conv2D(filters=filters,kernel_size=kernel_size,padding='same')(x)
return x
# 3 Dense block
def dense_block(tensor,k,reps):
'''
parameters:
k: filters
reps : repitition
run:
conv 1x1,4*k
conv 3x3,k
concatenate(tensor)
'''
for _ in range(reps):
x = bn_relu_conv(tensor,filters=4*k,kernel_size=1)
x = bn_relu_conv(x,filters=k,kernel_size=3)
tensor = Concatenate()([tensor,x])
return tensor
# 4 Transation block
def transition_layer(x,theta):
'''
parameters:
x: tensor
theta : compression
run:
bn_relu_conv 1x1,theta*k
AvgPool,p=2,s=2
'''
f = int(keras.backend.int_shape(x)[-1]*theta)
x = bn_relu_conv(x,f,kernel_size=1)
x = AvgPool2D(pool_size=2,strides=2,padding='same')(x)
return x
# 5 Mode
input = Input(shape=[224,224,3])
k = 32
theta = 0.5
repititions = 6,12,24,16
# 5.1 Conv(2*filters,k=7,s=2) + MaxPool(k=3,s=2)
x = Conv2D(filters=2*k ,kernel_size=7,strides=2,padding='same')(input)
x = MaxPool2D(pool_size=3,strides=2,padding='same')(x)
# 5.2 dense_block + transition_layer
for reps in repititions:
d = dense_block(x,k,reps)
x = transition_layer(d,theta)
# 5.3 GlobalAvgPool2D+ Dense
x = GlobalAvgPool2D()(d)
output = Dense(units=1000,activation='softmax')(x)
# 6
model = Model(inputs=input,outputs=output)
model.summary()
'''
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_1 (InputLayer) (None, 224, 224, 3) 0
__________________________________________________________________________________________________
conv2d_1 (Conv2D) (None, 112, 112, 64) 9472 input_1[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 56, 56, 64) 0 conv2d_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 56, 56, 64) 256 max_pooling2d_1[0][0]
__________________________________________________________________________________________________
re_lu_1 (ReLU) (None, 56, 56, 64) 0 batch_normalization_1[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D) (None, 56, 56, 128) 8320 re_lu_1[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 56, 56, 128) 512 conv2d_2[0][0]
__________________________________________________________________________________________________
re_lu_2 (ReLU) (None, 56, 56, 128) 0 batch_normalization_2[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D) (None, 56, 56, 32) 36896 re_lu_2[0][0]
__________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 56, 56, 96) 0 max_pooling2d_1[0][0]
conv2d_3[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 56, 56, 96) 384 concatenate_1[0][0]
__________________________________________________________________________________________________
re_lu_3 (ReLU) (None, 56, 56, 96) 0 batch_normalization_3[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D) (None, 56, 56, 128) 12416 re_lu_3[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 56, 56, 128) 512 conv2d_4[0][0]
__________________________________________________________________________________________________
re_lu_4 (ReLU) (None, 56, 56, 128) 0 batch_normalization_4[0][0]
__________________________________________________________________________________________________
conv2d_5 (Conv2D) (None, 56, 56, 32) 36896 re_lu_4[0][0]
__________________________________________________________________________________________________
concatenate_2 (Concatenate) (None, 56, 56, 128) 0 concatenate_1[0][0]
conv2d_5[0][0]
__________________________________________________________________________________________________
batch_normalization_5 (BatchNor (None, 56, 56, 128) 512 concatenate_2[0][0]
__________________________________________________________________________________________________
re_lu_5 (ReLU) (None, 56, 56, 128) 0 batch_normalization_5[0][0]
__________________________________________________________________________________________________
conv2d_6 (Conv2D) (None, 56, 56, 128) 16512 re_lu_5[0][0]
__________________________________________________________________________________________________
batch_normalization_6 (BatchNor (None, 56, 56, 128) 512 conv2d_6[0][0]
__________________________________________________________________________________________________
re_lu_6 (ReLU) (None, 56, 56, 128) 0 batch_normalization_6[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D) (None, 56, 56, 32) 36896 re_lu_6[0][0]
__________________________________________________________________________________________________
concatenate_3 (Concatenate) (None, 56, 56, 160) 0 concatenate_2[0][0]
conv2d_7[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 56, 56, 160) 640 concatenate_3[0][0]
__________________________________________________________________________________________________
re_lu_7 (ReLU) (None, 56, 56, 160) 0 batch_normalization_7[0][0]
__________________________________________________________________________________________________
conv2d_8 (Conv2D) (None, 56, 56, 128) 20608 re_lu_7[0][0]
__________________________________________________________________________________________________
batch_normalization_8 (BatchNor (None, 56, 56, 128) 512 conv2d_8[0][0]
__________________________________________________________________________________________________
re_lu_8 (ReLU) (None, 56, 56, 128) 0 batch_normalization_8[0][0]
__________________________________________________________________________________________________
conv2d_9 (Conv2D) (None, 56, 56, 32) 36896 re_lu_8[0][0]
__________________________________________________________________________________________________
concatenate_4 (Concatenate) (None, 56, 56, 192) 0 concatenate_3[0][0]
conv2d_9[0][0]
__________________________________________________________________________________________________
batch_normalization_9 (BatchNor (None, 56, 56, 192) 768 concatenate_4[0][0]
__________________________________________________________________________________________________
re_lu_9 (ReLU) (None, 56, 56, 192) 0 batch_normalization_9[0][0]
__________________________________________________________________________________________________
conv2d_10 (Conv2D) (None, 56, 56, 128) 24704 re_lu_9[0][0]
__________________________________________________________________________________________________
batch_normalization_10 (BatchNo (None, 56, 56, 128) 512 conv2d_10[0][0]
__________________________________________________________________________________________________
re_lu_10 (ReLU) (None, 56, 56, 128) 0 batch_normalization_10[0][0]
__________________________________________________________________________________________________
conv2d_11 (Conv2D) (None, 56, 56, 32) 36896 re_lu_10[0][0]
__________________________________________________________________________________________________
concatenate_5 (Concatenate) (None, 56, 56, 224) 0 concatenate_4[0][0]
conv2d_11[0][0]
__________________________________________________________________________________________________
batch_normalization_11 (BatchNo (None, 56, 56, 224) 896 concatenate_5[0][0]
__________________________________________________________________________________________________
re_lu_11 (ReLU) (None, 56, 56, 224) 0 batch_normalization_11[0][0]
__________________________________________________________________________________________________
conv2d_12 (Conv2D) (None, 56, 56, 128) 28800 re_lu_11[0][0]
__________________________________________________________________________________________________
batch_normalization_12 (BatchNo (None, 56, 56, 128) 512 conv2d_12[0][0]
__________________________________________________________________________________________________
re_lu_12 (ReLU) (None, 56, 56, 128) 0 batch_normalization_12[0][0]
__________________________________________________________________________________________________
conv2d_13 (Conv2D) (None, 56, 56, 32) 36896 re_lu_12[0][0]
__________________________________________________________________________________________________
concatenate_6 (Concatenate) (None, 56, 56, 256) 0 concatenate_5[0][0]
conv2d_13[0][0]
__________________________________________________________________________________________________
batch_normalization_13 (BatchNo (None, 56, 56, 256) 1024 concatenate_6[0][0]
__________________________________________________________________________________________________
re_lu_13 (ReLU) (None, 56, 56, 256) 0 batch_normalization_13[0][0]
__________________________________________________________________________________________________
conv2d_14 (Conv2D) (None, 56, 56, 128) 32896 re_lu_13[0][0]
__________________________________________________________________________________________________
average_pooling2d_1 (AveragePoo (None, 28, 28, 128) 0 conv2d_14[0][0]
__________________________________________________________________________________________________
batch_normalization_14 (BatchNo (None, 28, 28, 128) 512 average_pooling2d_1[0][0]
__________________________________________________________________________________________________
re_lu_14 (ReLU) (None, 28, 28, 128) 0 batch_normalization_14[0][0]
__________________________________________________________________________________________________
conv2d_15 (Conv2D) (None, 28, 28, 128) 16512 re_lu_14[0][0]
__________________________________________________________________________________________________
batch_normalization_15 (BatchNo (None, 28, 28, 128) 512 conv2d_15[0][0]
__________________________________________________________________________________________________
re_lu_15 (ReLU) (None, 28, 28, 128) 0 batch_normalization_15[0][0]
__________________________________________________________________________________________________
conv2d_16 (Conv2D) (None, 28, 28, 32) 36896 re_lu_15[0][0]
__________________________________________________________________________________________________
concatenate_7 (Concatenate) (None, 28, 28, 160) 0 average_pooling2d_1[0][0]
conv2d_16[0][0]
__________________________________________________________________________________________________
batch_normalization_16 (BatchNo (None, 28, 28, 160) 640 concatenate_7[0][0]
__________________________________________________________________________________________________
re_lu_16 (ReLU) (None, 28, 28, 160) 0 batch_normalization_16[0][0]
__________________________________________________________________________________________________
conv2d_17 (Conv2D) (None, 28, 28, 128) 20608 re_lu_16[0][0]
__________________________________________________________________________________________________
batch_normalization_17 (BatchNo (None, 28, 28, 128) 512 conv2d_17[0][0]
__________________________________________________________________________________________________
re_lu_17 (ReLU) (None, 28, 28, 128) 0 batch_normalization_17[0][0]
__________________________________________________________________________________________________
conv2d_18 (Conv2D) (None, 28, 28, 32) 36896 re_lu_17[0][0]
__________________________________________________________________________________________________
concatenate_8 (Concatenate) (None, 28, 28, 192) 0 concatenate_7[0][0]
conv2d_18[0][0]
__________________________________________________________________________________________________
batch_normalization_18 (BatchNo (None, 28, 28, 192) 768 concatenate_8[0][0]
__________________________________________________________________________________________________
re_lu_18 (ReLU) (None, 28, 28, 192) 0 batch_normalization_18[0][0]
__________________________________________________________________________________________________
conv2d_19 (Conv2D) (None, 28, 28, 128) 24704 re_lu_18[0][0]
__________________________________________________________________________________________________
batch_normalization_19 (BatchNo (None, 28, 28, 128) 512 conv2d_19[0][0]
__________________________________________________________________________________________________
re_lu_19 (ReLU) (None, 28, 28, 128) 0 batch_normalization_19[0][0]
__________________________________________________________________________________________________
conv2d_20 (Conv2D) (None, 28, 28, 32) 36896 re_lu_19[0][0]
__________________________________________________________________________________________________
concatenate_9 (Concatenate) (None, 28, 28, 224) 0 concatenate_8[0][0]
conv2d_20[0][0]
__________________________________________________________________________________________________
batch_normalization_20 (BatchNo (None, 28, 28, 224) 896 concatenate_9[0][0]
__________________________________________________________________________________________________
re_lu_20 (ReLU) (None, 28, 28, 224) 0 batch_normalization_20[0][0]
__________________________________________________________________________________________________
conv2d_21 (Conv2D) (None, 28, 28, 128) 28800 re_lu_20[0][0]
__________________________________________________________________________________________________
batch_normalization_21 (BatchNo (None, 28, 28, 128) 512 conv2d_21[0][0]
__________________________________________________________________________________________________
re_lu_21 (ReLU) (None, 28, 28, 128) 0 batch_normalization_21[0][0]
__________________________________________________________________________________________________
conv2d_22 (Conv2D) (None, 28, 28, 32) 36896 re_lu_21[0][0]
__________________________________________________________________________________________________
concatenate_10 (Concatenate) (None, 28, 28, 256) 0 concatenate_9[0][0]
conv2d_22[0][0]
__________________________________________________________________________________________________
batch_normalization_22 (BatchNo (None, 28, 28, 256) 1024 concatenate_10[0][0]
__________________________________________________________________________________________________
re_lu_22 (ReLU) (None, 28, 28, 256) 0 batch_normalization_22[0][0]
__________________________________________________________________________________________________
conv2d_23 (Conv2D) (None, 28, 28, 128) 32896 re_lu_22[0][0]
__________________________________________________________________________________________________
batch_normalization_23 (BatchNo (None, 28, 28, 128) 512 conv2d_23[0][0]
__________________________________________________________________________________________________
re_lu_23 (ReLU) (None, 28, 28, 128) 0 batch_normalization_23[0][0]
__________________________________________________________________________________________________
conv2d_24 (Conv2D) (None, 28, 28, 32) 36896 re_lu_23[0][0]
__________________________________________________________________________________________________
concatenate_11 (Concatenate) (None, 28, 28, 288) 0 concatenate_10[0][0]
conv2d_24[0][0]
__________________________________________________________________________________________________
batch_normalization_24 (BatchNo (None, 28, 28, 288) 1152 concatenate_11[0][0]
__________________________________________________________________________________________________
re_lu_24 (ReLU) (None, 28, 28, 288) 0 batch_normalization_24[0][0]
__________________________________________________________________________________________________
conv2d_25 (Conv2D) (None, 28, 28, 128) 36992 re_lu_24[0][0]
__________________________________________________________________________________________________
batch_normalization_25 (BatchNo (None, 28, 28, 128) 512 conv2d_25[0][0]
__________________________________________________________________________________________________
re_lu_25 (ReLU) (None, 28, 28, 128) 0 batch_normalization_25[0][0]
__________________________________________________________________________________________________
conv2d_26 (Conv2D) (None, 28, 28, 32) 36896 re_lu_25[0][0]
__________________________________________________________________________________________________
concatenate_12 (Concatenate) (None, 28, 28, 320) 0 concatenate_11[0][0]
conv2d_26[0][0]
__________________________________________________________________________________________________
batch_normalization_26 (BatchNo (None, 28, 28, 320) 1280 concatenate_12[0][0]
__________________________________________________________________________________________________
re_lu_26 (ReLU) (None, 28, 28, 320) 0 batch_normalization_26[0][0]
__________________________________________________________________________________________________
conv2d_27 (Conv2D) (None, 28, 28, 128) 41088 re_lu_26[0][0]
__________________________________________________________________________________________________
batch_normalization_27 (BatchNo (None, 28, 28, 128) 512 conv2d_27[0][0]
__________________________________________________________________________________________________
re_lu_27 (ReLU) (None, 28, 28, 128) 0 batch_normalization_27[0][0]
__________________________________________________________________________________________________
conv2d_28 (Conv2D) (None, 28, 28, 32) 36896 re_lu_27[0][0]
__________________________________________________________________________________________________
concatenate_13 (Concatenate) (None, 28, 28, 352) 0 concatenate_12[0][0]
conv2d_28[0][0]
__________________________________________________________________________________________________
batch_normalization_28 (BatchNo (None, 28, 28, 352) 1408 concatenate_13[0][0]
__________________________________________________________________________________________________
re_lu_28 (ReLU) (None, 28, 28, 352) 0 batch_normalization_28[0][0]
__________________________________________________________________________________________________
conv2d_29 (Conv2D) (None, 28, 28, 128) 45184 re_lu_28[0][0]
__________________________________________________________________________________________________
batch_normalization_29 (BatchNo (None, 28, 28, 128) 512 conv2d_29[0][0]
__________________________________________________________________________________________________
re_lu_29 (ReLU) (None, 28, 28, 128) 0 batch_normalization_29[0][0]
__________________________________________________________________________________________________
conv2d_30 (Conv2D) (None, 28, 28, 32) 36896 re_lu_29[0][0]
__________________________________________________________________________________________________
concatenate_14 (Concatenate) (None, 28, 28, 384) 0 concatenate_13[0][0]
conv2d_30[0][0]
__________________________________________________________________________________________________
batch_normalization_30 (BatchNo (None, 28, 28, 384) 1536 concatenate_14[0][0]
__________________________________________________________________________________________________
re_lu_30 (ReLU) (None, 28, 28, 384) 0 batch_normalization_30[0][0]
__________________________________________________________________________________________________
conv2d_31 (Conv2D) (None, 28, 28, 128) 49280 re_lu_30[0][0]
__________________________________________________________________________________________________
batch_normalization_31 (BatchNo (None, 28, 28, 128) 512 conv2d_31[0][0]
__________________________________________________________________________________________________
re_lu_31 (ReLU) (None, 28, 28, 128) 0 batch_normalization_31[0][0]
__________________________________________________________________________________________________
conv2d_32 (Conv2D) (None, 28, 28, 32) 36896 re_lu_31[0][0]
__________________________________________________________________________________________________
concatenate_15 (Concatenate) (None, 28, 28, 416) 0 concatenate_14[0][0]
conv2d_32[0][0]
__________________________________________________________________________________________________
batch_normalization_32 (BatchNo (None, 28, 28, 416) 1664 concatenate_15[0][0]
__________________________________________________________________________________________________
re_lu_32 (ReLU) (None, 28, 28, 416) 0 batch_normalization_32[0][0]
__________________________________________________________________________________________________
conv2d_33 (Conv2D) (None, 28, 28, 128) 53376 re_lu_32[0][0]
__________________________________________________________________________________________________
batch_normalization_33 (BatchNo (None, 28, 28, 128) 512 conv2d_33[0][0]
__________________________________________________________________________________________________
re_lu_33 (ReLU) (None, 28, 28, 128) 0 batch_normalization_33[0][0]
__________________________________________________________________________________________________
conv2d_34 (Conv2D) (None, 28, 28, 32) 36896 re_lu_33[0][0]
__________________________________________________________________________________________________
concatenate_16 (Concatenate) (None, 28, 28, 448) 0 concatenate_15[0][0]
conv2d_34[0][0]
__________________________________________________________________________________________________
batch_normalization_34 (BatchNo (None, 28, 28, 448) 1792 concatenate_16[0][0]
__________________________________________________________________________________________________
re_lu_34 (ReLU) (None, 28, 28, 448) 0 batch_normalization_34[0][0]
__________________________________________________________________________________________________
conv2d_35 (Conv2D) (None, 28, 28, 128) 57472 re_lu_34[0][0]
__________________________________________________________________________________________________
batch_normalization_35 (BatchNo (None, 28, 28, 128) 512 conv2d_35[0][0]
__________________________________________________________________________________________________
re_lu_35 (ReLU) (None, 28, 28, 128) 0 batch_normalization_35[0][0]
__________________________________________________________________________________________________
conv2d_36 (Conv2D) (None, 28, 28, 32) 36896 re_lu_35[0][0]
__________________________________________________________________________________________________
concatenate_17 (Concatenate) (None, 28, 28, 480) 0 concatenate_16[0][0]
conv2d_36[0][0]
__________________________________________________________________________________________________
batch_normalization_36 (BatchNo (None, 28, 28, 480) 1920 concatenate_17[0][0]
__________________________________________________________________________________________________
re_lu_36 (ReLU) (None, 28, 28, 480) 0 batch_normalization_36[0][0]
__________________________________________________________________________________________________
conv2d_37 (Conv2D) (None, 28, 28, 128) 61568 re_lu_36[0][0]
__________________________________________________________________________________________________
batch_normalization_37 (BatchNo (None, 28, 28, 128) 512 conv2d_37[0][0]
__________________________________________________________________________________________________
re_lu_37 (ReLU) (None, 28, 28, 128) 0 batch_normalization_37[0][0]
__________________________________________________________________________________________________
conv2d_38 (Conv2D) (None, 28, 28, 32) 36896 re_lu_37[0][0]
__________________________________________________________________________________________________
concatenate_18 (Concatenate) (None, 28, 28, 512) 0 concatenate_17[0][0]
conv2d_38[0][0]
__________________________________________________________________________________________________
batch_normalization_38 (BatchNo (None, 28, 28, 512) 2048 concatenate_18[0][0]
__________________________________________________________________________________________________
re_lu_38 (ReLU) (None, 28, 28, 512) 0 batch_normalization_38[0][0]
__________________________________________________________________________________________________
conv2d_39 (Conv2D) (None, 28, 28, 256) 131328 re_lu_38[0][0]
__________________________________________________________________________________________________
average_pooling2d_2 (AveragePoo (None, 14, 14, 256) 0 conv2d_39[0][0]
__________________________________________________________________________________________________
batch_normalization_39 (BatchNo (None, 14, 14, 256) 1024 average_pooling2d_2[0][0]
__________________________________________________________________________________________________
re_lu_39 (ReLU) (None, 14, 14, 256) 0 batch_normalization_39[0][0]
__________________________________________________________________________________________________
conv2d_40 (Conv2D) (None, 14, 14, 128) 32896 re_lu_39[0][0]
__________________________________________________________________________________________________
batch_normalization_40 (BatchNo (None, 14, 14, 128) 512 conv2d_40[0][0]
__________________________________________________________________________________________________
re_lu_40 (ReLU) (None, 14, 14, 128) 0 batch_normalization_40[0][0]
__________________________________________________________________________________________________
conv2d_41 (Conv2D) (None, 14, 14, 32) 36896 re_lu_40[0][0]
__________________________________________________________________________________________________
concatenate_19 (Concatenate) (None, 14, 14, 288) 0 average_pooling2d_2[0][0]
conv2d_41[0][0]
__________________________________________________________________________________________________
batch_normalization_41 (BatchNo (None, 14, 14, 288) 1152 concatenate_19[0][0]
__________________________________________________________________________________________________
re_lu_41 (ReLU) (None, 14, 14, 288) 0 batch_normalization_41[0][0]
__________________________________________________________________________________________________
conv2d_42 (Conv2D) (None, 14, 14, 128) 36992 re_lu_41[0][0]
__________________________________________________________________________________________________
batch_normalization_42 (BatchNo (None, 14, 14, 128) 512 conv2d_42[0][0]
__________________________________________________________________________________________________
re_lu_42 (ReLU) (None, 14, 14, 128) 0 batch_normalization_42[0][0]
__________________________________________________________________________________________________
conv2d_43 (Conv2D) (None, 14, 14, 32) 36896 re_lu_42[0][0]
__________________________________________________________________________________________________
concatenate_20 (Concatenate) (None, 14, 14, 320) 0 concatenate_19[0][0]
conv2d_43[0][0]
__________________________________________________________________________________________________
batch_normalization_43 (BatchNo (None, 14, 14, 320) 1280 concatenate_20[0][0]
__________________________________________________________________________________________________
re_lu_43 (ReLU) (None, 14, 14, 320) 0 batch_normalization_43[0][0]
__________________________________________________________________________________________________
conv2d_44 (Conv2D) (None, 14, 14, 128) 41088 re_lu_43[0][0]
__________________________________________________________________________________________________
batch_normalization_44 (BatchNo (None, 14, 14, 128) 512 conv2d_44[0][0]
__________________________________________________________________________________________________
re_lu_44 (ReLU) (None, 14, 14, 128) 0 batch_normalization_44[0][0]
__________________________________________________________________________________________________
conv2d_45 (Conv2D) (None, 14, 14, 32) 36896 re_lu_44[0][0]
__________________________________________________________________________________________________
concatenate_21 (Concatenate) (None, 14, 14, 352) 0 concatenate_20[0][0]
conv2d_45[0][0]
__________________________________________________________________________________________________
batch_normalization_45 (BatchNo (None, 14, 14, 352) 1408 concatenate_21[0][0]
__________________________________________________________________________________________________
re_lu_45 (ReLU) (None, 14, 14, 352) 0 batch_normalization_45[0][0]
__________________________________________________________________________________________________
conv2d_46 (Conv2D) (None, 14, 14, 128) 45184 re_lu_45[0][0]
__________________________________________________________________________________________________
batch_normalization_46 (BatchNo (None, 14, 14, 128) 512 conv2d_46[0][0]
__________________________________________________________________________________________________
re_lu_46 (ReLU) (None, 14, 14, 128) 0 batch_normalization_46[0][0]
__________________________________________________________________________________________________
conv2d_47 (Conv2D) (None, 14, 14, 32) 36896 re_lu_46[0][0]
__________________________________________________________________________________________________
concatenate_22 (Concatenate) (None, 14, 14, 384) 0 concatenate_21[0][0]
conv2d_47[0][0]
__________________________________________________________________________________________________
batch_normalization_47 (BatchNo (None, 14, 14, 384) 1536 concatenate_22[0][0]
__________________________________________________________________________________________________
re_lu_47 (ReLU) (None, 14, 14, 384) 0 batch_normalization_47[0][0]
__________________________________________________________________________________________________
conv2d_48 (Conv2D) (None, 14, 14, 128) 49280 re_lu_47[0][0]
__________________________________________________________________________________________________
batch_normalization_48 (BatchNo (None, 14, 14, 128) 512 conv2d_48[0][0]
__________________________________________________________________________________________________
re_lu_48 (ReLU) (None, 14, 14, 128) 0 batch_normalization_48[0][0]
__________________________________________________________________________________________________
conv2d_49 (Conv2D) (None, 14, 14, 32) 36896 re_lu_48[0][0]
__________________________________________________________________________________________________
concatenate_23 (Concatenate) (None, 14, 14, 416) 0 concatenate_22[0][0]
conv2d_49[0][0]
__________________________________________________________________________________________________
batch_normalization_49 (BatchNo (None, 14, 14, 416) 1664 concatenate_23[0][0]
__________________________________________________________________________________________________
re_lu_49 (ReLU) (None, 14, 14, 416) 0 batch_normalization_49[0][0]
__________________________________________________________________________________________________
conv2d_50 (Conv2D) (None, 14, 14, 128) 53376 re_lu_49[0][0]
__________________________________________________________________________________________________
batch_normalization_50 (BatchNo (None, 14, 14, 128) 512 conv2d_50[0][0]
__________________________________________________________________________________________________
re_lu_50 (ReLU) (None, 14, 14, 128) 0 batch_normalization_50[0][0]
__________________________________________________________________________________________________
conv2d_51 (Conv2D) (None, 14, 14, 32) 36896 re_lu_50[0][0]
__________________________________________________________________________________________________
concatenate_24 (Concatenate) (None, 14, 14, 448) 0 concatenate_23[0][0]
conv2d_51[0][0]
__________________________________________________________________________________________________
batch_normalization_51 (BatchNo (None, 14, 14, 448) 1792 concatenate_24[0][0]
__________________________________________________________________________________________________
re_lu_51 (ReLU) (None, 14, 14, 448) 0 batch_normalization_51[0][0]
__________________________________________________________________________________________________
conv2d_52 (Conv2D) (None, 14, 14, 128) 57472 re_lu_51[0][0]
__________________________________________________________________________________________________
batch_normalization_52 (BatchNo (None, 14, 14, 128) 512 conv2d_52[0][0]
__________________________________________________________________________________________________
re_lu_52 (ReLU) (None, 14, 14, 128) 0 batch_normalization_52[0][0]
__________________________________________________________________________________________________
conv2d_53 (Conv2D) (None, 14, 14, 32) 36896 re_lu_52[0][0]
__________________________________________________________________________________________________
concatenate_25 (Concatenate) (None, 14, 14, 480) 0 concatenate_24[0][0]
conv2d_53[0][0]
__________________________________________________________________________________________________
batch_normalization_53 (BatchNo (None, 14, 14, 480) 1920 concatenate_25[0][0]
__________________________________________________________________________________________________
re_lu_53 (ReLU) (None, 14, 14, 480) 0 batch_normalization_53[0][0]
__________________________________________________________________________________________________
conv2d_54 (Conv2D) (None, 14, 14, 128) 61568 re_lu_53[0][0]
__________________________________________________________________________________________________
batch_normalization_54 (BatchNo (None, 14, 14, 128) 512 conv2d_54[0][0]
__________________________________________________________________________________________________
re_lu_54 (ReLU) (None, 14, 14, 128) 0 batch_normalization_54[0][0]
__________________________________________________________________________________________________
conv2d_55 (Conv2D) (None, 14, 14, 32) 36896 re_lu_54[0][0]
__________________________________________________________________________________________________
concatenate_26 (Concatenate) (None, 14, 14, 512) 0 concatenate_25[0][0]
conv2d_55[0][0]
__________________________________________________________________________________________________
batch_normalization_55 (BatchNo (None, 14, 14, 512) 2048 concatenate_26[0][0]
__________________________________________________________________________________________________
re_lu_55 (ReLU) (None, 14, 14, 512) 0 batch_normalization_55[0][0]
__________________________________________________________________________________________________
conv2d_56 (Conv2D) (None, 14, 14, 128) 65664 re_lu_55[0][0]
__________________________________________________________________________________________________
batch_normalization_56 (BatchNo (None, 14, 14, 128) 512 conv2d_56[0][0]
__________________________________________________________________________________________________
re_lu_56 (ReLU) (None, 14, 14, 128) 0 batch_normalization_56[0][0]
__________________________________________________________________________________________________
conv2d_57 (Conv2D) (None, 14, 14, 32) 36896 re_lu_56[0][0]
__________________________________________________________________________________________________
concatenate_27 (Concatenate) (None, 14, 14, 544) 0 concatenate_26[0][0]
conv2d_57[0][0]
__________________________________________________________________________________________________
batch_normalization_57 (BatchNo (None, 14, 14, 544) 2176 concatenate_27[0][0]
__________________________________________________________________________________________________
re_lu_57 (ReLU) (None, 14, 14, 544) 0 batch_normalization_57[0][0]
__________________________________________________________________________________________________
conv2d_58 (Conv2D) (None, 14, 14, 128) 69760 re_lu_57[0][0]
__________________________________________________________________________________________________
batch_normalization_58 (BatchNo (None, 14, 14, 128) 512 conv2d_58[0][0]
__________________________________________________________________________________________________
re_lu_58 (ReLU) (None, 14, 14, 128) 0 batch_normalization_58[0][0]
__________________________________________________________________________________________________
conv2d_59 (Conv2D) (None, 14, 14, 32) 36896 re_lu_58[0][0]
__________________________________________________________________________________________________
concatenate_28 (Concatenate) (None, 14, 14, 576) 0 concatenate_27[0][0]
conv2d_59[0][0]
__________________________________________________________________________________________________
batch_normalization_59 (BatchNo (None, 14, 14, 576) 2304 concatenate_28[0][0]
__________________________________________________________________________________________________
re_lu_59 (ReLU) (None, 14, 14, 576) 0 batch_normalization_59[0][0]
__________________________________________________________________________________________________
conv2d_60 (Conv2D) (None, 14, 14, 128) 73856 re_lu_59[0][0]
__________________________________________________________________________________________________
batch_normalization_60 (BatchNo (None, 14, 14, 128) 512 conv2d_60[0][0]
__________________________________________________________________________________________________
re_lu_60 (ReLU) (None, 14, 14, 128) 0 batch_normalization_60[0][0]
__________________________________________________________________________________________________
conv2d_61 (Conv2D) (None, 14, 14, 32) 36896 re_lu_60[0][0]
__________________________________________________________________________________________________
concatenate_29 (Concatenate) (None, 14, 14, 608) 0 concatenate_28[0][0]
conv2d_61[0][0]
__________________________________________________________________________________________________
batch_normalization_61 (BatchNo (None, 14, 14, 608) 2432 concatenate_29[0][0]
__________________________________________________________________________________________________
re_lu_61 (ReLU) (None, 14, 14, 608) 0 batch_normalization_61[0][0]
__________________________________________________________________________________________________
conv2d_62 (Conv2D) (None, 14, 14, 128) 77952 re_lu_61[0][0]
__________________________________________________________________________________________________
batch_normalization_62 (BatchNo (None, 14, 14, 128) 512 conv2d_62[0][0]
__________________________________________________________________________________________________
re_lu_62 (ReLU) (None, 14, 14, 128) 0 batch_normalization_62[0][0]
__________________________________________________________________________________________________
conv2d_63 (Conv2D) (None, 14, 14, 32) 36896 re_lu_62[0][0]
__________________________________________________________________________________________________
concatenate_30 (Concatenate) (None, 14, 14, 640) 0 concatenate_29[0][0]
conv2d_63[0][0]
__________________________________________________________________________________________________
batch_normalization_63 (BatchNo (None, 14, 14, 640) 2560 concatenate_30[0][0]
__________________________________________________________________________________________________
re_lu_63 (ReLU) (None, 14, 14, 640) 0 batch_normalization_63[0][0]
__________________________________________________________________________________________________
conv2d_64 (Conv2D) (None, 14, 14, 128) 82048 re_lu_63[0][0]
__________________________________________________________________________________________________
batch_normalization_64 (BatchNo (None, 14, 14, 128) 512 conv2d_64[0][0]
__________________________________________________________________________________________________
re_lu_64 (ReLU) (None, 14, 14, 128) 0 batch_normalization_64[0][0]
__________________________________________________________________________________________________
conv2d_65 (Conv2D) (None, 14, 14, 32) 36896 re_lu_64[0][0]
__________________________________________________________________________________________________
concatenate_31 (Concatenate) (None, 14, 14, 672) 0 concatenate_30[0][0]
conv2d_65[0][0]
__________________________________________________________________________________________________
batch_normalization_65 (BatchNo (None, 14, 14, 672) 2688 concatenate_31[0][0]
__________________________________________________________________________________________________
re_lu_65 (ReLU) (None, 14, 14, 672) 0 batch_normalization_65[0][0]
__________________________________________________________________________________________________
conv2d_66 (Conv2D) (None, 14, 14, 128) 86144 re_lu_65[0][0]
__________________________________________________________________________________________________
batch_normalization_66 (BatchNo (None, 14, 14, 128) 512 conv2d_66[0][0]
__________________________________________________________________________________________________
re_lu_66 (ReLU) (None, 14, 14, 128) 0 batch_normalization_66[0][0]
__________________________________________________________________________________________________
conv2d_67 (Conv2D) (None, 14, 14, 32) 36896 re_lu_66[0][0]
__________________________________________________________________________________________________
concatenate_32 (Concatenate) (None, 14, 14, 704) 0 concatenate_31[0][0]
conv2d_67[0][0]
__________________________________________________________________________________________________
batch_normalization_67 (BatchNo (None, 14, 14, 704) 2816 concatenate_32[0][0]
__________________________________________________________________________________________________
re_lu_67 (ReLU) (None, 14, 14, 704) 0 batch_normalization_67[0][0]
__________________________________________________________________________________________________
conv2d_68 (Conv2D) (None, 14, 14, 128) 90240 re_lu_67[0][0]
__________________________________________________________________________________________________
batch_normalization_68 (BatchNo (None, 14, 14, 128) 512 conv2d_68[0][0]
__________________________________________________________________________________________________
re_lu_68 (ReLU) (None, 14, 14, 128) 0 batch_normalization_68[0][0]
__________________________________________________________________________________________________
conv2d_69 (Conv2D) (None, 14, 14, 32) 36896 re_lu_68[0][0]
__________________________________________________________________________________________________
concatenate_33 (Concatenate) (None, 14, 14, 736) 0 concatenate_32[0][0]
conv2d_69[0][0]
__________________________________________________________________________________________________
batch_normalization_69 (BatchNo (None, 14, 14, 736) 2944 concatenate_33[0][0]
__________________________________________________________________________________________________
re_lu_69 (ReLU) (None, 14, 14, 736) 0 batch_normalization_69[0][0]
__________________________________________________________________________________________________
conv2d_70 (Conv2D) (None, 14, 14, 128) 94336 re_lu_69[0][0]
__________________________________________________________________________________________________
batch_normalization_70 (BatchNo (None, 14, 14, 128) 512 conv2d_70[0][0]
__________________________________________________________________________________________________
re_lu_70 (ReLU) (None, 14, 14, 128) 0 batch_normalization_70[0][0]
__________________________________________________________________________________________________
conv2d_71 (Conv2D) (None, 14, 14, 32) 36896 re_lu_70[0][0]
__________________________________________________________________________________________________
concatenate_34 (Concatenate) (None, 14, 14, 768) 0 concatenate_33[0][0]
conv2d_71[0][0]
__________________________________________________________________________________________________
batch_normalization_71 (BatchNo (None, 14, 14, 768) 3072 concatenate_34[0][0]
__________________________________________________________________________________________________
re_lu_71 (ReLU) (None, 14, 14, 768) 0 batch_normalization_71[0][0]
__________________________________________________________________________________________________
conv2d_72 (Conv2D) (None, 14, 14, 128) 98432 re_lu_71[0][0]
__________________________________________________________________________________________________
batch_normalization_72 (BatchNo (None, 14, 14, 128) 512 conv2d_72[0][0]
__________________________________________________________________________________________________
re_lu_72 (ReLU) (None, 14, 14, 128) 0 batch_normalization_72[0][0]
__________________________________________________________________________________________________
conv2d_73 (Conv2D) (None, 14, 14, 32) 36896 re_lu_72[0][0]
__________________________________________________________________________________________________
concatenate_35 (Concatenate) (None, 14, 14, 800) 0 concatenate_34[0][0]
conv2d_73[0][0]
__________________________________________________________________________________________________
batch_normalization_73 (BatchNo (None, 14, 14, 800) 3200 concatenate_35[0][0]
__________________________________________________________________________________________________
re_lu_73 (ReLU) (None, 14, 14, 800) 0 batch_normalization_73[0][0]
__________________________________________________________________________________________________
conv2d_74 (Conv2D) (None, 14, 14, 128) 102528 re_lu_73[0][0]
__________________________________________________________________________________________________
batch_normalization_74 (BatchNo (None, 14, 14, 128) 512 conv2d_74[0][0]
__________________________________________________________________________________________________
re_lu_74 (ReLU) (None, 14, 14, 128) 0 batch_normalization_74[0][0]
__________________________________________________________________________________________________
conv2d_75 (Conv2D) (None, 14, 14, 32) 36896 re_lu_74[0][0]
__________________________________________________________________________________________________
concatenate_36 (Concatenate) (None, 14, 14, 832) 0 concatenate_35[0][0]
conv2d_75[0][0]
__________________________________________________________________________________________________
batch_normalization_75 (BatchNo (None, 14, 14, 832) 3328 concatenate_36[0][0]
__________________________________________________________________________________________________
re_lu_75 (ReLU) (None, 14, 14, 832) 0 batch_normalization_75[0][0]
__________________________________________________________________________________________________
conv2d_76 (Conv2D) (None, 14, 14, 128) 106624 re_lu_75[0][0]
__________________________________________________________________________________________________
batch_normalization_76 (BatchNo (None, 14, 14, 128) 512 conv2d_76[0][0]
__________________________________________________________________________________________________
re_lu_76 (ReLU) (None, 14, 14, 128) 0 batch_normalization_76[0][0]
__________________________________________________________________________________________________
conv2d_77 (Conv2D) (None, 14, 14, 32) 36896 re_lu_76[0][0]
__________________________________________________________________________________________________
concatenate_37 (Concatenate) (None, 14, 14, 864) 0 concatenate_36[0][0]
conv2d_77[0][0]
__________________________________________________________________________________________________
batch_normalization_77 (BatchNo (None, 14, 14, 864) 3456 concatenate_37[0][0]
__________________________________________________________________________________________________
re_lu_77 (ReLU) (None, 14, 14, 864) 0 batch_normalization_77[0][0]
__________________________________________________________________________________________________
conv2d_78 (Conv2D) (None, 14, 14, 128) 110720 re_lu_77[0][0]
__________________________________________________________________________________________________
batch_normalization_78 (BatchNo (None, 14, 14, 128) 512 conv2d_78[0][0]
__________________________________________________________________________________________________
re_lu_78 (ReLU) (None, 14, 14, 128) 0 batch_normalization_78[0][0]
__________________________________________________________________________________________________
conv2d_79 (Conv2D) (None, 14, 14, 32) 36896 re_lu_78[0][0]
__________________________________________________________________________________________________
concatenate_38 (Concatenate) (None, 14, 14, 896) 0 concatenate_37[0][0]
conv2d_79[0][0]
__________________________________________________________________________________________________
batch_normalization_79 (BatchNo (None, 14, 14, 896) 3584 concatenate_38[0][0]
__________________________________________________________________________________________________
re_lu_79 (ReLU) (None, 14, 14, 896) 0 batch_normalization_79[0][0]
__________________________________________________________________________________________________
conv2d_80 (Conv2D) (None, 14, 14, 128) 114816 re_lu_79[0][0]
__________________________________________________________________________________________________
batch_normalization_80 (BatchNo (None, 14, 14, 128) 512 conv2d_80[0][0]
__________________________________________________________________________________________________
re_lu_80 (ReLU) (None, 14, 14, 128) 0 batch_normalization_80[0][0]
__________________________________________________________________________________________________
conv2d_81 (Conv2D) (None, 14, 14, 32) 36896 re_lu_80[0][0]
__________________________________________________________________________________________________
concatenate_39 (Concatenate) (None, 14, 14, 928) 0 concatenate_38[0][0]
conv2d_81[0][0]
__________________________________________________________________________________________________
batch_normalization_81 (BatchNo (None, 14, 14, 928) 3712 concatenate_39[0][0]
__________________________________________________________________________________________________
re_lu_81 (ReLU) (None, 14, 14, 928) 0 batch_normalization_81[0][0]
__________________________________________________________________________________________________
conv2d_82 (Conv2D) (None, 14, 14, 128) 118912 re_lu_81[0][0]
__________________________________________________________________________________________________
batch_normalization_82 (BatchNo (None, 14, 14, 128) 512 conv2d_82[0][0]
__________________________________________________________________________________________________
re_lu_82 (ReLU) (None, 14, 14, 128) 0 batch_normalization_82[0][0]
__________________________________________________________________________________________________
conv2d_83 (Conv2D) (None, 14, 14, 32) 36896 re_lu_82[0][0]
__________________________________________________________________________________________________
concatenate_40 (Concatenate) (None, 14, 14, 960) 0 concatenate_39[0][0]
conv2d_83[0][0]
__________________________________________________________________________________________________
batch_normalization_83 (BatchNo (None, 14, 14, 960) 3840 concatenate_40[0][0]
__________________________________________________________________________________________________
re_lu_83 (ReLU) (None, 14, 14, 960) 0 batch_normalization_83[0][0]
__________________________________________________________________________________________________
conv2d_84 (Conv2D) (None, 14, 14, 128) 123008 re_lu_83[0][0]
__________________________________________________________________________________________________
batch_normalization_84 (BatchNo (None, 14, 14, 128) 512 conv2d_84[0][0]
__________________________________________________________________________________________________
re_lu_84 (ReLU) (None, 14, 14, 128) 0 batch_normalization_84[0][0]
__________________________________________________________________________________________________
conv2d_85 (Conv2D) (None, 14, 14, 32) 36896 re_lu_84[0][0]
__________________________________________________________________________________________________
concatenate_41 (Concatenate) (None, 14, 14, 992) 0 concatenate_40[0][0]
conv2d_85[0][0]
__________________________________________________________________________________________________
batch_normalization_85 (BatchNo (None, 14, 14, 992) 3968 concatenate_41[0][0]
__________________________________________________________________________________________________
re_lu_85 (ReLU) (None, 14, 14, 992) 0 batch_normalization_85[0][0]
__________________________________________________________________________________________________
conv2d_86 (Conv2D) (None, 14, 14, 128) 127104 re_lu_85[0][0]
__________________________________________________________________________________________________
batch_normalization_86 (BatchNo (None, 14, 14, 128) 512 conv2d_86[0][0]
__________________________________________________________________________________________________
re_lu_86 (ReLU) (None, 14, 14, 128) 0 batch_normalization_86[0][0]
__________________________________________________________________________________________________
conv2d_87 (Conv2D) (None, 14, 14, 32) 36896 re_lu_86[0][0]
__________________________________________________________________________________________________
concatenate_42 (Concatenate) (None, 14, 14, 1024) 0 concatenate_41[0][0]
conv2d_87[0][0]
__________________________________________________________________________________________________
batch_normalization_87 (BatchNo (None, 14, 14, 1024) 4096 concatenate_42[0][0]
__________________________________________________________________________________________________
re_lu_87 (ReLU) (None, 14, 14, 1024) 0 batch_normalization_87[0][0]
__________________________________________________________________________________________________
conv2d_88 (Conv2D) (None, 14, 14, 512) 524800 re_lu_87[0][0]
__________________________________________________________________________________________________
average_pooling2d_3 (AveragePoo (None, 7, 7, 512) 0 conv2d_88[0][0]
__________________________________________________________________________________________________
batch_normalization_88 (BatchNo (None, 7, 7, 512) 2048 average_pooling2d_3[0][0]
__________________________________________________________________________________________________
re_lu_88 (ReLU) (None, 7, 7, 512) 0 batch_normalization_88[0][0]
__________________________________________________________________________________________________
conv2d_89 (Conv2D) (None, 7, 7, 128) 65664 re_lu_88[0][0]
__________________________________________________________________________________________________
batch_normalization_89 (BatchNo (None, 7, 7, 128) 512 conv2d_89[0][0]
__________________________________________________________________________________________________
re_lu_89 (ReLU) (None, 7, 7, 128) 0 batch_normalization_89[0][0]
__________________________________________________________________________________________________
conv2d_90 (Conv2D) (None, 7, 7, 32) 36896 re_lu_89[0][0]
__________________________________________________________________________________________________
concatenate_43 (Concatenate) (None, 7, 7, 544) 0 average_pooling2d_3[0][0]
conv2d_90[0][0]
__________________________________________________________________________________________________
batch_normalization_90 (BatchNo (None, 7, 7, 544) 2176 concatenate_43[0][0]
__________________________________________________________________________________________________
re_lu_90 (ReLU) (None, 7, 7, 544) 0 batch_normalization_90[0][0]
__________________________________________________________________________________________________
conv2d_91 (Conv2D) (None, 7, 7, 128) 69760 re_lu_90[0][0]
__________________________________________________________________________________________________
batch_normalization_91 (BatchNo (None, 7, 7, 128) 512 conv2d_91[0][0]
__________________________________________________________________________________________________
re_lu_91 (ReLU) (None, 7, 7, 128) 0 batch_normalization_91[0][0]
__________________________________________________________________________________________________
conv2d_92 (Conv2D) (None, 7, 7, 32) 36896 re_lu_91[0][0]
__________________________________________________________________________________________________
concatenate_44 (Concatenate) (None, 7, 7, 576) 0 concatenate_43[0][0]
conv2d_92[0][0]
__________________________________________________________________________________________________
batch_normalization_92 (BatchNo (None, 7, 7, 576) 2304 concatenate_44[0][0]
__________________________________________________________________________________________________
re_lu_92 (ReLU) (None, 7, 7, 576) 0 batch_normalization_92[0][0]
__________________________________________________________________________________________________
conv2d_93 (Conv2D) (None, 7, 7, 128) 73856 re_lu_92[0][0]
__________________________________________________________________________________________________
batch_normalization_93 (BatchNo (None, 7, 7, 128) 512 conv2d_93[0][0]
__________________________________________________________________________________________________
re_lu_93 (ReLU) (None, 7, 7, 128) 0 batch_normalization_93[0][0]
__________________________________________________________________________________________________
conv2d_94 (Conv2D) (None, 7, 7, 32) 36896 re_lu_93[0][0]
__________________________________________________________________________________________________
concatenate_45 (Concatenate) (None, 7, 7, 608) 0 concatenate_44[0][0]
conv2d_94[0][0]
__________________________________________________________________________________________________
batch_normalization_94 (BatchNo (None, 7, 7, 608) 2432 concatenate_45[0][0]
__________________________________________________________________________________________________
re_lu_94 (ReLU) (None, 7, 7, 608) 0 batch_normalization_94[0][0]
__________________________________________________________________________________________________
conv2d_95 (Conv2D) (None, 7, 7, 128) 77952 re_lu_94[0][0]
__________________________________________________________________________________________________
batch_normalization_95 (BatchNo (None, 7, 7, 128) 512 conv2d_95[0][0]
__________________________________________________________________________________________________
re_lu_95 (ReLU) (None, 7, 7, 128) 0 batch_normalization_95[0][0]
__________________________________________________________________________________________________
conv2d_96 (Conv2D) (None, 7, 7, 32) 36896 re_lu_95[0][0]
__________________________________________________________________________________________________
concatenate_46 (Concatenate) (None, 7, 7, 640) 0 concatenate_45[0][0]
conv2d_96[0][0]
__________________________________________________________________________________________________
batch_normalization_96 (BatchNo (None, 7, 7, 640) 2560 concatenate_46[0][0]
__________________________________________________________________________________________________
re_lu_96 (ReLU) (None, 7, 7, 640) 0 batch_normalization_96[0][0]
__________________________________________________________________________________________________
conv2d_97 (Conv2D) (None, 7, 7, 128) 82048 re_lu_96[0][0]
__________________________________________________________________________________________________
batch_normalization_97 (BatchNo (None, 7, 7, 128) 512 conv2d_97[0][0]
__________________________________________________________________________________________________
re_lu_97 (ReLU) (None, 7, 7, 128) 0 batch_normalization_97[0][0]
__________________________________________________________________________________________________
conv2d_98 (Conv2D) (None, 7, 7, 32) 36896 re_lu_97[0][0]
__________________________________________________________________________________________________
concatenate_47 (Concatenate) (None, 7, 7, 672) 0 concatenate_46[0][0]
conv2d_98[0][0]
__________________________________________________________________________________________________
batch_normalization_98 (BatchNo (None, 7, 7, 672) 2688 concatenate_47[0][0]
__________________________________________________________________________________________________
re_lu_98 (ReLU) (None, 7, 7, 672) 0 batch_normalization_98[0][0]
__________________________________________________________________________________________________
conv2d_99 (Conv2D) (None, 7, 7, 128) 86144 re_lu_98[0][0]
__________________________________________________________________________________________________
batch_normalization_99 (BatchNo (None, 7, 7, 128) 512 conv2d_99[0][0]
__________________________________________________________________________________________________
re_lu_99 (ReLU) (None, 7, 7, 128) 0 batch_normalization_99[0][0]
__________________________________________________________________________________________________
conv2d_100 (Conv2D) (None, 7, 7, 32) 36896 re_lu_99[0][0]
__________________________________________________________________________________________________
concatenate_48 (Concatenate) (None, 7, 7, 704) 0 concatenate_47[0][0]
conv2d_100[0][0]
__________________________________________________________________________________________________
batch_normalization_100 (BatchN (None, 7, 7, 704) 2816 concatenate_48[0][0]
__________________________________________________________________________________________________
re_lu_100 (ReLU) (None, 7, 7, 704) 0 batch_normalization_100[0][0]
__________________________________________________________________________________________________
conv2d_101 (Conv2D) (None, 7, 7, 128) 90240 re_lu_100[0][0]
__________________________________________________________________________________________________
batch_normalization_101 (BatchN (None, 7, 7, 128) 512 conv2d_101[0][0]
__________________________________________________________________________________________________
re_lu_101 (ReLU) (None, 7, 7, 128) 0 batch_normalization_101[0][0]
__________________________________________________________________________________________________
conv2d_102 (Conv2D) (None, 7, 7, 32) 36896 re_lu_101[0][0]
__________________________________________________________________________________________________
concatenate_49 (Concatenate) (None, 7, 7, 736) 0 concatenate_48[0][0]
conv2d_102[0][0]
__________________________________________________________________________________________________
batch_normalization_102 (BatchN (None, 7, 7, 736) 2944 concatenate_49[0][0]
__________________________________________________________________________________________________
re_lu_102 (ReLU) (None, 7, 7, 736) 0 batch_normalization_102[0][0]
__________________________________________________________________________________________________
conv2d_103 (Conv2D) (None, 7, 7, 128) 94336 re_lu_102[0][0]
__________________________________________________________________________________________________
batch_normalization_103 (BatchN (None, 7, 7, 128) 512 conv2d_103[0][0]
__________________________________________________________________________________________________
re_lu_103 (ReLU) (None, 7, 7, 128) 0 batch_normalization_103[0][0]
__________________________________________________________________________________________________
conv2d_104 (Conv2D) (None, 7, 7, 32) 36896 re_lu_103[0][0]
__________________________________________________________________________________________________
concatenate_50 (Concatenate) (None, 7, 7, 768) 0 concatenate_49[0][0]
conv2d_104[0][0]
__________________________________________________________________________________________________
batch_normalization_104 (BatchN (None, 7, 7, 768) 3072 concatenate_50[0][0]
__________________________________________________________________________________________________
re_lu_104 (ReLU) (None, 7, 7, 768) 0 batch_normalization_104[0][0]
__________________________________________________________________________________________________
conv2d_105 (Conv2D) (None, 7, 7, 128) 98432 re_lu_104[0][0]
__________________________________________________________________________________________________
batch_normalization_105 (BatchN (None, 7, 7, 128) 512 conv2d_105[0][0]
__________________________________________________________________________________________________
re_lu_105 (ReLU) (None, 7, 7, 128) 0 batch_normalization_105[0][0]
__________________________________________________________________________________________________
conv2d_106 (Conv2D) (None, 7, 7, 32) 36896 re_lu_105[0][0]
__________________________________________________________________________________________________
concatenate_51 (Concatenate) (None, 7, 7, 800) 0 concatenate_50[0][0]
conv2d_106[0][0]
__________________________________________________________________________________________________
batch_normalization_106 (BatchN (None, 7, 7, 800) 3200 concatenate_51[0][0]
__________________________________________________________________________________________________
re_lu_106 (ReLU) (None, 7, 7, 800) 0 batch_normalization_106[0][0]
__________________________________________________________________________________________________
conv2d_107 (Conv2D) (None, 7, 7, 128) 102528 re_lu_106[0][0]
__________________________________________________________________________________________________
batch_normalization_107 (BatchN (None, 7, 7, 128) 512 conv2d_107[0][0]
__________________________________________________________________________________________________
re_lu_107 (ReLU) (None, 7, 7, 128) 0 batch_normalization_107[0][0]
__________________________________________________________________________________________________
conv2d_108 (Conv2D) (None, 7, 7, 32) 36896 re_lu_107[0][0]
__________________________________________________________________________________________________
concatenate_52 (Concatenate) (None, 7, 7, 832) 0 concatenate_51[0][0]
conv2d_108[0][0]
__________________________________________________________________________________________________
batch_normalization_108 (BatchN (None, 7, 7, 832) 3328 concatenate_52[0][0]
__________________________________________________________________________________________________
re_lu_108 (ReLU) (None, 7, 7, 832) 0 batch_normalization_108[0][0]
__________________________________________________________________________________________________
conv2d_109 (Conv2D) (None, 7, 7, 128) 106624 re_lu_108[0][0]
__________________________________________________________________________________________________
batch_normalization_109 (BatchN (None, 7, 7, 128) 512 conv2d_109[0][0]
__________________________________________________________________________________________________
re_lu_109 (ReLU) (None, 7, 7, 128) 0 batch_normalization_109[0][0]
__________________________________________________________________________________________________
conv2d_110 (Conv2D) (None, 7, 7, 32) 36896 re_lu_109[0][0]
__________________________________________________________________________________________________
concatenate_53 (Concatenate) (None, 7, 7, 864) 0 concatenate_52[0][0]
conv2d_110[0][0]
__________________________________________________________________________________________________
batch_normalization_110 (BatchN (None, 7, 7, 864) 3456 concatenate_53[0][0]
__________________________________________________________________________________________________
re_lu_110 (ReLU) (None, 7, 7, 864) 0 batch_normalization_110[0][0]
__________________________________________________________________________________________________
conv2d_111 (Conv2D) (None, 7, 7, 128) 110720 re_lu_110[0][0]
__________________________________________________________________________________________________
batch_normalization_111 (BatchN (None, 7, 7, 128) 512 conv2d_111[0][0]
__________________________________________________________________________________________________
re_lu_111 (ReLU) (None, 7, 7, 128) 0 batch_normalization_111[0][0]
__________________________________________________________________________________________________
conv2d_112 (Conv2D) (None, 7, 7, 32) 36896 re_lu_111[0][0]
__________________________________________________________________________________________________
concatenate_54 (Concatenate) (None, 7, 7, 896) 0 concatenate_53[0][0]
conv2d_112[0][0]
__________________________________________________________________________________________________
batch_normalization_112 (BatchN (None, 7, 7, 896) 3584 concatenate_54[0][0]
__________________________________________________________________________________________________
re_lu_112 (ReLU) (None, 7, 7, 896) 0 batch_normalization_112[0][0]
__________________________________________________________________________________________________
conv2d_113 (Conv2D) (None, 7, 7, 128) 114816 re_lu_112[0][0]
__________________________________________________________________________________________________
batch_normalization_113 (BatchN (None, 7, 7, 128) 512 conv2d_113[0][0]
__________________________________________________________________________________________________
re_lu_113 (ReLU) (None, 7, 7, 128) 0 batch_normalization_113[0][0]
__________________________________________________________________________________________________
conv2d_114 (Conv2D) (None, 7, 7, 32) 36896 re_lu_113[0][0]
__________________________________________________________________________________________________
concatenate_55 (Concatenate) (None, 7, 7, 928) 0 concatenate_54[0][0]
conv2d_114[0][0]
__________________________________________________________________________________________________
batch_normalization_114 (BatchN (None, 7, 7, 928) 3712 concatenate_55[0][0]
__________________________________________________________________________________________________
re_lu_114 (ReLU) (None, 7, 7, 928) 0 batch_normalization_114[0][0]
__________________________________________________________________________________________________
conv2d_115 (Conv2D) (None, 7, 7, 128) 118912 re_lu_114[0][0]
__________________________________________________________________________________________________
batch_normalization_115 (BatchN (None, 7, 7, 128) 512 conv2d_115[0][0]
__________________________________________________________________________________________________
re_lu_115 (ReLU) (None, 7, 7, 128) 0 batch_normalization_115[0][0]
__________________________________________________________________________________________________
conv2d_116 (Conv2D) (None, 7, 7, 32) 36896 re_lu_115[0][0]
__________________________________________________________________________________________________
concatenate_56 (Concatenate) (None, 7, 7, 960) 0 concatenate_55[0][0]
conv2d_116[0][0]
__________________________________________________________________________________________________
batch_normalization_116 (BatchN (None, 7, 7, 960) 3840 concatenate_56[0][0]
__________________________________________________________________________________________________
re_lu_116 (ReLU) (None, 7, 7, 960) 0 batch_normalization_116[0][0]
__________________________________________________________________________________________________
conv2d_117 (Conv2D) (None, 7, 7, 128) 123008 re_lu_116[0][0]
__________________________________________________________________________________________________
batch_normalization_117 (BatchN (None, 7, 7, 128) 512 conv2d_117[0][0]
__________________________________________________________________________________________________
re_lu_117 (ReLU) (None, 7, 7, 128) 0 batch_normalization_117[0][0]
__________________________________________________________________________________________________
conv2d_118 (Conv2D) (None, 7, 7, 32) 36896 re_lu_117[0][0]
__________________________________________________________________________________________________
concatenate_57 (Concatenate) (None, 7, 7, 992) 0 concatenate_56[0][0]
conv2d_118[0][0]
__________________________________________________________________________________________________
batch_normalization_118 (BatchN (None, 7, 7, 992) 3968 concatenate_57[0][0]
__________________________________________________________________________________________________
re_lu_118 (ReLU) (None, 7, 7, 992) 0 batch_normalization_118[0][0]
__________________________________________________________________________________________________
conv2d_119 (Conv2D) (None, 7, 7, 128) 127104 re_lu_118[0][0]
__________________________________________________________________________________________________
batch_normalization_119 (BatchN (None, 7, 7, 128) 512 conv2d_119[0][0]
__________________________________________________________________________________________________
re_lu_119 (ReLU) (None, 7, 7, 128) 0 batch_normalization_119[0][0]
__________________________________________________________________________________________________
conv2d_120 (Conv2D) (None, 7, 7, 32) 36896 re_lu_119[0][0]
__________________________________________________________________________________________________
concatenate_58 (Concatenate) (None, 7, 7, 1024) 0 concatenate_57[0][0]
conv2d_120[0][0]
__________________________________________________________________________________________________
global_average_pooling2d_1 (Glo (None, 1024) 0 concatenate_58[0][0]
__________________________________________________________________________________________________
dense_1 (Dense) (None, 1000) 1025000 global_average_pooling2d_1[0][0]
==================================================================================================
Total params: 8,068,392
Trainable params: 7,986,920
Non-trainable params: 81,472
__________________________________________________________________________________________________
Process finished with exit code 0
'''