InceptionV3的PyTorch实现:https://github.com/pytorch/vision/blob/master/torchvision/models/inception.py
2a表示第2组的第1个Block,同一组的空间维度相同
但为何没有3a, 5a?
(299, 299, 3)
→【1a, Cout=32, f=3, s=2】→(149, 149, 32)
→【2a, Cout=32, f=3】→(147, 147, 32)→【2b, Cout=64, f=3, p=1】→(147, 147, 64)
→【max pool, f=3, s=2】→(73, 73, 64)
→【3b, Cout=80, f=1】→(73, 73, 80)
→【4a, Cout=192, f=3】→(71, 71, 192)→【max pool, f=3, s=2】→(35, 35, 192)
→【Mixed_5b, InceptionA】→(35, 35, 256)→【Mixed_5c, InceptionA】→(35, 35, 288)
→【Mixed_5d, InceptionA】→(35, 35, 288)
→【Mixed_6a, InceptionB】→(17, 17, 768)
→【Mixed_6b, InceptionC】→(17, 17, 768)→【Mixed_6c, InceptionC】→(17, 17, 768)
→【Mixed_6d, InceptionC】→(17, 17, 768)→【Mixed_6e, InceptionC】→(17, 17, 768)
→【Mixed_7a, InceptionD】→(8, 8, 1280)
→【Mixed_7b, InceptionE】→(8, 8, 2048)→【Mixed_7c, InceptionE】→(8, 8, 2048)
→【global avg pool】→(2048,)→【dropout】→(2048,)→【fc】→(1000,)
分支:→【Mixed_6e, InceptionC】→(17, 17, 768)→【InceptionAux】→(1000,)
InceptionA
使用了3次,分别用在Mixed_5b, Mixed_5c, Mixed_5d
中,包含参数pool_features
,输入为(35, 35, in_channels)
,输出固定为(35, 35, 224+pool features)
(35, 35, 192)→【Mixed_5b,InceptionA, pool_features=32】→(35, 35, 224+32=256)
(35, 35, 256)→【Mixed_5c,InceptionA, pool_features=64】→(35, 35, 224+64=288)
(35, 35, 288)→【Mixed_5d,InceptionA, pool_features=64】→(35, 35, 224+64=288)
以(35, 35, 192)→【Mixed_5b, InceptionA, pool_features=32】→(35, 35, 256)
为例
输入:(35, 35, 192)
分支1:→【BasicConv2d, Cout=64, f=1】→(35, 35, 64)
分支2:→【BasicConv2d, Cout=48, f=1】→(35, 35, 48)
→【BasicConv2d, Cout=64, f=5, p=2】→(35, 35, 64)
分支3:→【BasicConv2d, Cout=64, f=1】→(35, 35, 64)
→【BasicConv2d, Cout=96, f=3, p=1】→(35, 35, 96)
→【BasicConv2d, Cout=96, f=3, p=1】→(35, 35, 96)
分支4:→【avg pool, f=3, s=1, p=1】→(35, 35, 192)
→【BasicConv2d, Cout=pool_features, f=1】→(35, 35, pool_features)
合并:(35, 35, 224+pool_features)
InceptionB
只使用了1次,用在Mixed_6a
中
(35, 35, 288)→【Mixed_6a, InceptionB】→(17, 17, 768)
,空间维度减半,通道数增加到大约2.7倍
输入:(35, 35, 288)
分支1:→【BasicConv2d, Cout=384, f=3, s=2】→(17, 17, 384)
分支2:→【BasicConv2d, Cout=64, f=1】→(35, 35, 64)
→【BasicConv2d, Cout=96, f=3, p=1】→(35, 35, 96)
→【BasicConv2d, Cout=96, f=3, s=2】→(17, 17, 96)
分支3:→【max pool, f=3, s=2】→(17, 17, 288)
合并:(17, 17, 384+96+288=768)
InceptionC
使用了4次,分别用在Mixed_6b, Mixed_6c, Mixed_6d, Mixed_6e
中,输入和输出均为(17, 17, 768)
,只是参数channels_7x7
不同,参数channels_7x7
简记为c7
Mixed_6b,c7=128
Mixed_6c,c7=160
Mixed_6d,c7=160
Mixed_6e,c7=192
输入:(17, 17, 768)
分支1:→【BasicConv2d, Cout=192, f=1】→(17, 17, 192)
分支2:→【BasicConv2d, Cout=c7, f=1】→(17, 17, c7)
→【BasicConv2d, Cout=c7, f=(1, 7), p=(0, 3)】→(17, 17, c7)
→【BasicConv2d, Cout=192, f=(7, 1), p=(3, 0)】→(17, 17, 192)
分支3:→【BasicConv2d, Cout=c7, f=1】→(17, 17, c7)
→【BasicConv2d, Cout=c7, f=(7, 1), p=(3, 0)】→(17, 17, c7)
→【BasicConv2d, Cout=c7, f=(1, 7), p=(0, 3)】→(17, 17, c7)
→【BasicConv2d, Cout=c7, f=(7, 1), p=(3, 0)】→(17, 17, c7)
→【BasicConv2d, Cout=192, f=(1, 7), p=(0, 3)】→(17, 17, 192)
分支4:→【avg pool, f=3, s=1, p=1】→(17, 17, 768)
→【BasicConv2d, Cout=192, f=1】→(17, 17, 192)
合并:(17, 17, 192×4=768)
InceptionD
只使用了1次,用在Mixed_7a
中
(17, 17, 768)→【Mixed_7a, InceptionD】→(8, 8, 1280)
,空间维度减半,通道数增加到大约1.7倍
输入:(17, 17, 768)
分支1:→【BasicConv2d, Cout=192, f=1】→(17, 17, 192)
→【BasicConv2d, Cout=320, f=3, s=2】→(8, 8, 320)
分支2:→【BasicConv2d, Cout=192, f=1】→(17, 17, 192)
→【BasicConv2d, Cout=192, f=(1, 7), p=(0, 3)】→(17, 17, 192)
→【BasicConv2d, Cout=192, f=(7, 1), p=(3, 0)】→(17, 17, 192)
→【BasicConv2d, Cout=192, f=3, s=2】→(8, 8, 192)
分支3:→【max pool, f=3, s=2】→(8, 8, 768)
合并:(8, 8, 320+192+768=1280)
InceptionE
使用了2次, 分别用在Mixed_7b, Mixed_7c
中,输入为(8, 8, in_channels)
,输出固定为(8, 8, 2048)
以(8, 8, 1280)→【Mixed_7b, InceptionE】→(8, 8, 2048)
为例
输入:(8, 8, 1280)
分支1:→【BasicConv2d, Cout=320, f=1】→(8, 8, 320)
分支2:→【BasicConv2d, Cout=384, f=1】→(8, 8, 384)
分支2-1:→【BasicConv2d, Cout=384, f=(1, 3), p=(0, 1)】→(8, 8, 384)
分支2-2:→【BasicConv2d, Cout=384, f=(3, 1), p=(1, 0)】→(8, 8, 384)
合并:(8, 8, 384×2=768)
分支3:→【BasicConv2d, Cout=448, f=1】→(8, 8, 448)
→【BasicConv2d, Cout=384, f=3, p=1】→(8, 8, 384)
分支3-1:→【BasicConv2d, Cout=384, f=(1, 3), p=(0, 1)】→(8, 8, 384)
分支3-2:→【BasicConv2d, Cout=384, f=(3, 1), p=(1, 0)】→(8, 8, 384)
合并:(8, 8, 384×2=768)
分支4:→【avg pool, f=3, s=1, p=1】→(8, 8, 1280)
→【BasicConv2d, Cout=192, f=1】→(8, 8, 192)
合并:(8, 8, 320+768+768+192=2048)
InceptionAux
只用了1次,连接在【Mixed_6e, InceptionC】
的输出(17, 17, 768)
上
(17, 17, 768)
→【avg pool, f=5, s=3】→(5, 5, 768)
→【BasicConv2d, Cout=128, f=1】→(5, 5, 128)
→【BasicConv2d, Cout=768, f=5】→(1, 1, 768)
→【reshape】→(768,)→【fc】→(1000,)
pool层几个值得注意的地方
(147, 147, 64)→【max pool, f=3, s=2】→(73, 73, 64)
(71, 71, 192)→【max pool, f=3, s=2】→(35, 35, 192),使用f=3(一般使用f=2)
Keras中的InceptionV3,共313层
参考:https://github.com/keras-team/keras-applications/blob/master/keras_applications/inception_v3.py
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
第0层
__________________________________________________________________________________________________
input_1 (InputLayer) (None, 299, 299, 3) 0
__________________________________________________________________________________________________
第1-3层,(299, 299, 3)→【1a, Cout=32, f=3, s=2】→(149, 149, 32)
__________________________________________________________________________________________________
conv2d_1 (Conv2D) (None, 149, 149, 32) 864 input_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 149, 149, 32) 96 conv2d_1[0][0]
__________________________________________________________________________________________________
activation_1 (Activation) (None, 149, 149, 32) 0 batch_normalization_1[0][0]
__________________________________________________________________________________________________
第4-6层,(149, 149, 32)→【2a, Cout=32, f=3】→(147, 147, 32)
__________________________________________________________________________________________________
conv2d_2 (Conv2D) (None, 147, 147, 32) 9216 activation_1[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 147, 147, 32) 96 conv2d_2[0][0]
__________________________________________________________________________________________________
activation_2 (Activation) (None, 147, 147, 32) 0 batch_normalization_2[0][0]
__________________________________________________________________________________________________
第7-9层,(147, 147, 32)→【2b, Cout=64, f=3, p=1】→(147, 147, 64)
__________________________________________________________________________________________________
conv2d_3 (Conv2D) (None, 147, 147, 64) 18432 activation_2[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 147, 147, 64) 192 conv2d_3[0][0]
__________________________________________________________________________________________________
activation_3 (Activation) (None, 147, 147, 64) 0 batch_normalization_3[0][0]
__________________________________________________________________________________________________
第10层,(147, 147, 64)→【max pool, f=3, s=2】→(73, 73, 64)
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 73, 73, 64) 0 activation_3[0][0]
__________________________________________________________________________________________________
第11-13层,(73, 73, 64)→【3b, Cout=80, f=1】→(73, 73, 80)
__________________________________________________________________________________________________
conv2d_4 (Conv2D) (None, 73, 73, 80) 5120 max_pooling2d_1[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 73, 73, 80) 240 conv2d_4[0][0]
__________________________________________________________________________________________________
activation_4 (Activation) (None, 73, 73, 80) 0 batch_normalization_4[0][0]
__________________________________________________________________________________________________
第14-16层,(73, 73, 80)→【4a, Cout=192, f=3】→(71, 71, 192)
__________________________________________________________________________________________________
conv2d_5 (Conv2D) (None, 71, 71, 192) 138240 activation_4[0][0]
__________________________________________________________________________________________________
batch_normalization_5 (BatchNor (None, 71, 71, 192) 576 conv2d_5[0][0]
__________________________________________________________________________________________________
activation_5 (Activation) (None, 71, 71, 192) 0 batch_normalization_5[0][0]
__________________________________________________________________________________________________
第17层,(71, 71, 192)→【max pool, f=3, s=2】→(35, 35, 192)
__________________________________________________________________________________________________
max_pooling2d_2 (MaxPooling2D) (None, 35, 35, 192) 0 activation_5[0][0]
__________________________________________________________________________________________________
第18-40层,(35, 35, 192)→【Mixed_5b, InceptionA】→(35, 35, 256)
__________________________________________________________________________________________________
conv2d_9 (Conv2D) (None, 35, 35, 64) 12288 max_pooling2d_2[0][0]
__________________________________________________________________________________________________
batch_normalization_9 (BatchNor (None, 35, 35, 64) 192 conv2d_9[0][0]
__________________________________________________________________________________________________
activation_9 (Activation) (None, 35, 35, 64) 0 batch_normalization_9[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D) (None, 35, 35, 48) 9216 max_pooling2d_2[0][0]
__________________________________________________________________________________________________
conv2d_10 (Conv2D) (None, 35, 35, 96) 55296 activation_9[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 35, 35, 48) 144 conv2d_7[0][0]
__________________________________________________________________________________________________
batch_normalization_10 (BatchNo (None, 35, 35, 96) 288 conv2d_10[0][0]
__________________________________________________________________________________________________
activation_7 (Activation) (None, 35, 35, 48) 0 batch_normalization_7[0][0]
__________________________________________________________________________________________________
activation_10 (Activation) (None, 35, 35, 96) 0 batch_normalization_10[0][0]
__________________________________________________________________________________________________
average_pooling2d_1 (AveragePoo (None, 35, 35, 192) 0 max_pooling2d_2[0][0]
__________________________________________________________________________________________________
conv2d_6 (Conv2D) (None, 35, 35, 64) 12288 max_pooling2d_2[0][0]
__________________________________________________________________________________________________
conv2d_8 (Conv2D) (None, 35, 35, 64) 76800 activation_7[0][0]
__________________________________________________________________________________________________
conv2d_11 (Conv2D) (None, 35, 35, 96) 82944 activation_10[0][0]
__________________________________________________________________________________________________
conv2d_12 (Conv2D) (None, 35, 35, 32) 6144 average_pooling2d_1[0][0]
__________________________________________________________________________________________________
batch_normalization_6 (BatchNor (None, 35, 35, 64) 192 conv2d_6[0][0]
__________________________________________________________________________________________________
batch_normalization_8 (BatchNor (None, 35, 35, 64) 192 conv2d_8[0][0]
__________________________________________________________________________________________________
batch_normalization_11 (BatchNo (None, 35, 35, 96) 288 conv2d_11[0][0]
__________________________________________________________________________________________________
batch_normalization_12 (BatchNo (None, 35, 35, 32) 96 conv2d_12[0][0]
__________________________________________________________________________________________________
activation_6 (Activation) (None, 35, 35, 64) 0 batch_normalization_6[0][0]
__________________________________________________________________________________________________
activation_8 (Activation) (None, 35, 35, 64) 0 batch_normalization_8[0][0]
__________________________________________________________________________________________________
activation_11 (Activation) (None, 35, 35, 96) 0 batch_normalization_11[0][0]
__________________________________________________________________________________________________
activation_12 (Activation) (None, 35, 35, 32) 0 batch_normalization_12[0][0]
__________________________________________________________________________________________________
mixed0 (Concatenate) (None, 35, 35, 256) 0 activation_6[0][0]
activation_8[0][0]
activation_11[0][0]
activation_12[0][0]
__________________________________________________________________________________________________
第41-63层,(35, 35, 256)→【Mixed_5c, InceptionA】→(35, 35, 288)
__________________________________________________________________________________________________
conv2d_16 (Conv2D) (None, 35, 35, 64) 16384 mixed0[0][0]
__________________________________________________________________________________________________
batch_normalization_16 (BatchNo (None, 35, 35, 64) 192 conv2d_16[0][0]
__________________________________________________________________________________________________
activation_16 (Activation) (None, 35, 35, 64) 0 batch_normalization_16[0][0]
__________________________________________________________________________________________________
conv2d_14 (Conv2D) (None, 35, 35, 48) 12288 mixed0[0][0]
__________________________________________________________________________________________________
conv2d_17 (Conv2D) (None, 35, 35, 96) 55296 activation_16[0][0]
__________________________________________________________________________________________________
batch_normalization_14 (BatchNo (None, 35, 35, 48) 144 conv2d_14[0][0]
__________________________________________________________________________________________________
batch_normalization_17 (BatchNo (None, 35, 35, 96) 288 conv2d_17[0][0]
__________________________________________________________________________________________________
activation_14 (Activation) (None, 35, 35, 48) 0 batch_normalization_14[0][0]
__________________________________________________________________________________________________
activation_17 (Activation) (None, 35, 35, 96) 0 batch_normalization_17[0][0]
__________________________________________________________________________________________________
average_pooling2d_2 (AveragePoo (None, 35, 35, 256) 0 mixed0[0][0]
__________________________________________________________________________________________________
conv2d_13 (Conv2D) (None, 35, 35, 64) 16384 mixed0[0][0]
__________________________________________________________________________________________________
conv2d_15 (Conv2D) (None, 35, 35, 64) 76800 activation_14[0][0]
__________________________________________________________________________________________________
conv2d_18 (Conv2D) (None, 35, 35, 96) 82944 activation_17[0][0]
__________________________________________________________________________________________________
conv2d_19 (Conv2D) (None, 35, 35, 64) 16384 average_pooling2d_2[0][0]
__________________________________________________________________________________________________
batch_normalization_13 (BatchNo (None, 35, 35, 64) 192 conv2d_13[0][0]
__________________________________________________________________________________________________
batch_normalization_15 (BatchNo (None, 35, 35, 64) 192 conv2d_15[0][0]
__________________________________________________________________________________________________
batch_normalization_18 (BatchNo (None, 35, 35, 96) 288 conv2d_18[0][0]
__________________________________________________________________________________________________
batch_normalization_19 (BatchNo (None, 35, 35, 64) 192 conv2d_19[0][0]
__________________________________________________________________________________________________
activation_13 (Activation) (None, 35, 35, 64) 0 batch_normalization_13[0][0]
__________________________________________________________________________________________________
activation_15 (Activation) (None, 35, 35, 64) 0 batch_normalization_15[0][0]
__________________________________________________________________________________________________
activation_18 (Activation) (None, 35, 35, 96) 0 batch_normalization_18[0][0]
__________________________________________________________________________________________________
activation_19 (Activation) (None, 35, 35, 64) 0 batch_normalization_19[0][0]
__________________________________________________________________________________________________
mixed1 (Concatenate) (None, 35, 35, 288) 0 activation_13[0][0]
activation_15[0][0]
activation_18[0][0]
activation_19[0][0]
__________________________________________________________________________________________________
第64-86层,(35, 35, 288)→【Mixed_5d, InceptionA】→(35, 35, 288)
__________________________________________________________________________________________________
conv2d_23 (Conv2D) (None, 35, 35, 64) 18432 mixed1[0][0]
__________________________________________________________________________________________________
batch_normalization_23 (BatchNo (None, 35, 35, 64) 192 conv2d_23[0][0]
__________________________________________________________________________________________________
activation_23 (Activation) (None, 35, 35, 64) 0 batch_normalization_23[0][0]
__________________________________________________________________________________________________
conv2d_21 (Conv2D) (None, 35, 35, 48) 13824 mixed1[0][0]
__________________________________________________________________________________________________
conv2d_24 (Conv2D) (None, 35, 35, 96) 55296 activation_23[0][0]
__________________________________________________________________________________________________
batch_normalization_21 (BatchNo (None, 35, 35, 48) 144 conv2d_21[0][0]
__________________________________________________________________________________________________
batch_normalization_24 (BatchNo (None, 35, 35, 96) 288 conv2d_24[0][0]
__________________________________________________________________________________________________
activation_21 (Activation) (None, 35, 35, 48) 0 batch_normalization_21[0][0]
__________________________________________________________________________________________________
activation_24 (Activation) (None, 35, 35, 96) 0 batch_normalization_24[0][0]
__________________________________________________________________________________________________
average_pooling2d_3 (AveragePoo (None, 35, 35, 288) 0 mixed1[0][0]
__________________________________________________________________________________________________
conv2d_20 (Conv2D) (None, 35, 35, 64) 18432 mixed1[0][0]
__________________________________________________________________________________________________
conv2d_22 (Conv2D) (None, 35, 35, 64) 76800 activation_21[0][0]
__________________________________________________________________________________________________
conv2d_25 (Conv2D) (None, 35, 35, 96) 82944 activation_24[0][0]
__________________________________________________________________________________________________
conv2d_26 (Conv2D) (None, 35, 35, 64) 18432 average_pooling2d_3[0][0]
__________________________________________________________________________________________________
batch_normalization_20 (BatchNo (None, 35, 35, 64) 192 conv2d_20[0][0]
__________________________________________________________________________________________________
batch_normalization_22 (BatchNo (None, 35, 35, 64) 192 conv2d_22[0][0]
__________________________________________________________________________________________________
batch_normalization_25 (BatchNo (None, 35, 35, 96) 288 conv2d_25[0][0]
__________________________________________________________________________________________________
batch_normalization_26 (BatchNo (None, 35, 35, 64) 192 conv2d_26[0][0]
__________________________________________________________________________________________________
activation_20 (Activation) (None, 35, 35, 64) 0 batch_normalization_20[0][0]
__________________________________________________________________________________________________
activation_22 (Activation) (None, 35, 35, 64) 0 batch_normalization_22[0][0]
__________________________________________________________________________________________________
activation_25 (Activation) (None, 35, 35, 96) 0 batch_normalization_25[0][0]
__________________________________________________________________________________________________
activation_26 (Activation) (None, 35, 35, 64) 0 batch_normalization_26[0][0]
__________________________________________________________________________________________________
mixed2 (Concatenate) (None, 35, 35, 288) 0 activation_20[0][0]
activation_22[0][0]
activation_25[0][0]
activation_26[0][0]
__________________________________________________________________________________________________
第87-100层,(35, 35, 288)→【Mixed_6a, InceptionB】→(17, 17, 768)
__________________________________________________________________________________________________
conv2d_28 (Conv2D) (None, 35, 35, 64) 18432 mixed2[0][0]
__________________________________________________________________________________________________
batch_normalization_28 (BatchNo (None, 35, 35, 64) 192 conv2d_28[0][0]
__________________________________________________________________________________________________
activation_28 (Activation) (None, 35, 35, 64) 0 batch_normalization_28[0][0]
__________________________________________________________________________________________________
conv2d_29 (Conv2D) (None, 35, 35, 96) 55296 activation_28[0][0]
__________________________________________________________________________________________________
batch_normalization_29 (BatchNo (None, 35, 35, 96) 288 conv2d_29[0][0]
__________________________________________________________________________________________________
activation_29 (Activation) (None, 35, 35, 96) 0 batch_normalization_29[0][0]
__________________________________________________________________________________________________
conv2d_27 (Conv2D) (None, 17, 17, 384) 995328 mixed2[0][0]
__________________________________________________________________________________________________
conv2d_30 (Conv2D) (None, 17, 17, 96) 82944 activation_29[0][0]
__________________________________________________________________________________________________
batch_normalization_27 (BatchNo (None, 17, 17, 384) 1152 conv2d_27[0][0]
__________________________________________________________________________________________________
batch_normalization_30 (BatchNo (None, 17, 17, 96) 288 conv2d_30[0][0]
__________________________________________________________________________________________________
activation_27 (Activation) (None, 17, 17, 384) 0 batch_normalization_27[0][0]
__________________________________________________________________________________________________
activation_30 (Activation) (None, 17, 17, 96) 0 batch_normalization_30[0][0]
__________________________________________________________________________________________________
max_pooling2d_3 (MaxPooling2D) (None, 17, 17, 288) 0 mixed2[0][0]
__________________________________________________________________________________________________
mixed3 (Concatenate) (None, 17, 17, 768) 0 activation_27[0][0]
activation_30[0][0]
max_pooling2d_3[0][0]
__________________________________________________________________________________________________
第101-132层,(17, 17, 768)→【Mixed_6b, InceptionC】→(17, 17, 768)
__________________________________________________________________________________________________
conv2d_35 (Conv2D) (None, 17, 17, 128) 98304 mixed3[0][0]
__________________________________________________________________________________________________
batch_normalization_35 (BatchNo (None, 17, 17, 128) 384 conv2d_35[0][0]
__________________________________________________________________________________________________
activation_35 (Activation) (None, 17, 17, 128) 0 batch_normalization_35[0][0]
__________________________________________________________________________________________________
conv2d_36 (Conv2D) (None, 17, 17, 128) 114688 activation_35[0][0]
__________________________________________________________________________________________________
batch_normalization_36 (BatchNo (None, 17, 17, 128) 384 conv2d_36[0][0]
__________________________________________________________________________________________________
activation_36 (Activation) (None, 17, 17, 128) 0 batch_normalization_36[0][0]
__________________________________________________________________________________________________
conv2d_32 (Conv2D) (None, 17, 17, 128) 98304 mixed3[0][0]
__________________________________________________________________________________________________
conv2d_37 (Conv2D) (None, 17, 17, 128) 114688 activation_36[0][0]
__________________________________________________________________________________________________
batch_normalization_32 (BatchNo (None, 17, 17, 128) 384 conv2d_32[0][0]
__________________________________________________________________________________________________
batch_normalization_37 (BatchNo (None, 17, 17, 128) 384 conv2d_37[0][0]
__________________________________________________________________________________________________
activation_32 (Activation) (None, 17, 17, 128) 0 batch_normalization_32[0][0]
__________________________________________________________________________________________________
activation_37 (Activation) (None, 17, 17, 128) 0 batch_normalization_37[0][0]
__________________________________________________________________________________________________
conv2d_33 (Conv2D) (None, 17, 17, 128) 114688 activation_32[0][0]
__________________________________________________________________________________________________
conv2d_38 (Conv2D) (None, 17, 17, 128) 114688 activation_37[0][0]
__________________________________________________________________________________________________
batch_normalization_33 (BatchNo (None, 17, 17, 128) 384 conv2d_33[0][0]
__________________________________________________________________________________________________
batch_normalization_38 (BatchNo (None, 17, 17, 128) 384 conv2d_38[0][0]
__________________________________________________________________________________________________
activation_33 (Activation) (None, 17, 17, 128) 0 batch_normalization_33[0][0]
__________________________________________________________________________________________________
activation_38 (Activation) (None, 17, 17, 128) 0 batch_normalization_38[0][0]
__________________________________________________________________________________________________
average_pooling2d_4 (AveragePoo (None, 17, 17, 768) 0 mixed3[0][0]
__________________________________________________________________________________________________
conv2d_31 (Conv2D) (None, 17, 17, 192) 147456 mixed3[0][0]
__________________________________________________________________________________________________
conv2d_34 (Conv2D) (None, 17, 17, 192) 172032 activation_33[0][0]
__________________________________________________________________________________________________
conv2d_39 (Conv2D) (None, 17, 17, 192) 172032 activation_38[0][0]
__________________________________________________________________________________________________
conv2d_40 (Conv2D) (None, 17, 17, 192) 147456 average_pooling2d_4[0][0]
__________________________________________________________________________________________________
batch_normalization_31 (BatchNo (None, 17, 17, 192) 576 conv2d_31[0][0]
__________________________________________________________________________________________________
batch_normalization_34 (BatchNo (None, 17, 17, 192) 576 conv2d_34[0][0]
__________________________________________________________________________________________________
batch_normalization_39 (BatchNo (None, 17, 17, 192) 576 conv2d_39[0][0]
__________________________________________________________________________________________________
batch_normalization_40 (BatchNo (None, 17, 17, 192) 576 conv2d_40[0][0]
__________________________________________________________________________________________________
activation_31 (Activation) (None, 17, 17, 192) 0 batch_normalization_31[0][0]
__________________________________________________________________________________________________
activation_34 (Activation) (None, 17, 17, 192) 0 batch_normalization_34[0][0]
__________________________________________________________________________________________________
activation_39 (Activation) (None, 17, 17, 192) 0 batch_normalization_39[0][0]
__________________________________________________________________________________________________
activation_40 (Activation) (None, 17, 17, 192) 0 batch_normalization_40[0][0]
__________________________________________________________________________________________________
mixed4 (Concatenate) (None, 17, 17, 768) 0 activation_31[0][0]
activation_34[0][0]
activation_39[0][0]
activation_40[0][0]
__________________________________________________________________________________________________
第133-164层,(17, 17, 768)→【Mixed_6c, InceptionC】→(17, 17, 768)
__________________________________________________________________________________________________
conv2d_45 (Conv2D) (None, 17, 17, 160) 122880 mixed4[0][0]
__________________________________________________________________________________________________
batch_normalization_45 (BatchNo (None, 17, 17, 160) 480 conv2d_45[0][0]
__________________________________________________________________________________________________
activation_45 (Activation) (None, 17, 17, 160) 0 batch_normalization_45[0][0]
__________________________________________________________________________________________________
conv2d_46 (Conv2D) (None, 17, 17, 160) 179200 activation_45[0][0]
__________________________________________________________________________________________________
batch_normalization_46 (BatchNo (None, 17, 17, 160) 480 conv2d_46[0][0]
__________________________________________________________________________________________________
activation_46 (Activation) (None, 17, 17, 160) 0 batch_normalization_46[0][0]
__________________________________________________________________________________________________
conv2d_42 (Conv2D) (None, 17, 17, 160) 122880 mixed4[0][0]
__________________________________________________________________________________________________
conv2d_47 (Conv2D) (None, 17, 17, 160) 179200 activation_46[0][0]
__________________________________________________________________________________________________
batch_normalization_42 (BatchNo (None, 17, 17, 160) 480 conv2d_42[0][0]
__________________________________________________________________________________________________
batch_normalization_47 (BatchNo (None, 17, 17, 160) 480 conv2d_47[0][0]
__________________________________________________________________________________________________
activation_42 (Activation) (None, 17, 17, 160) 0 batch_normalization_42[0][0]
__________________________________________________________________________________________________
activation_47 (Activation) (None, 17, 17, 160) 0 batch_normalization_47[0][0]
__________________________________________________________________________________________________
conv2d_43 (Conv2D) (None, 17, 17, 160) 179200 activation_42[0][0]
__________________________________________________________________________________________________
conv2d_48 (Conv2D) (None, 17, 17, 160) 179200 activation_47[0][0]
__________________________________________________________________________________________________
batch_normalization_43 (BatchNo (None, 17, 17, 160) 480 conv2d_43[0][0]
__________________________________________________________________________________________________
batch_normalization_48 (BatchNo (None, 17, 17, 160) 480 conv2d_48[0][0]
__________________________________________________________________________________________________
activation_43 (Activation) (None, 17, 17, 160) 0 batch_normalization_43[0][0]
__________________________________________________________________________________________________
activation_48 (Activation) (None, 17, 17, 160) 0 batch_normalization_48[0][0]
__________________________________________________________________________________________________
average_pooling2d_5 (AveragePoo (None, 17, 17, 768) 0 mixed4[0][0]
__________________________________________________________________________________________________
conv2d_41 (Conv2D) (None, 17, 17, 192) 147456 mixed4[0][0]
__________________________________________________________________________________________________
conv2d_44 (Conv2D) (None, 17, 17, 192) 215040 activation_43[0][0]
__________________________________________________________________________________________________
conv2d_49 (Conv2D) (None, 17, 17, 192) 215040 activation_48[0][0]
__________________________________________________________________________________________________
conv2d_50 (Conv2D) (None, 17, 17, 192) 147456 average_pooling2d_5[0][0]
__________________________________________________________________________________________________
batch_normalization_41 (BatchNo (None, 17, 17, 192) 576 conv2d_41[0][0]
__________________________________________________________________________________________________
batch_normalization_44 (BatchNo (None, 17, 17, 192) 576 conv2d_44[0][0]
__________________________________________________________________________________________________
batch_normalization_49 (BatchNo (None, 17, 17, 192) 576 conv2d_49[0][0]
__________________________________________________________________________________________________
batch_normalization_50 (BatchNo (None, 17, 17, 192) 576 conv2d_50[0][0]
__________________________________________________________________________________________________
activation_41 (Activation) (None, 17, 17, 192) 0 batch_normalization_41[0][0]
__________________________________________________________________________________________________
activation_44 (Activation) (None, 17, 17, 192) 0 batch_normalization_44[0][0]
__________________________________________________________________________________________________
activation_49 (Activation) (None, 17, 17, 192) 0 batch_normalization_49[0][0]
__________________________________________________________________________________________________
activation_50 (Activation) (None, 17, 17, 192) 0 batch_normalization_50[0][0]
__________________________________________________________________________________________________
mixed5 (Concatenate) (None, 17, 17, 768) 0 activation_41[0][0]
activation_44[0][0]
activation_49[0][0]
activation_50[0][0]
__________________________________________________________________________________________________
第165-196层,(17, 17, 768)→【Mixed_6d, InceptionC】→(17, 17, 768)
__________________________________________________________________________________________________
conv2d_55 (Conv2D) (None, 17, 17, 160) 122880 mixed5[0][0]
__________________________________________________________________________________________________
batch_normalization_55 (BatchNo (None, 17, 17, 160) 480 conv2d_55[0][0]
__________________________________________________________________________________________________
activation_55 (Activation) (None, 17, 17, 160) 0 batch_normalization_55[0][0]
__________________________________________________________________________________________________
conv2d_56 (Conv2D) (None, 17, 17, 160) 179200 activation_55[0][0]
__________________________________________________________________________________________________
batch_normalization_56 (BatchNo (None, 17, 17, 160) 480 conv2d_56[0][0]
__________________________________________________________________________________________________
activation_56 (Activation) (None, 17, 17, 160) 0 batch_normalization_56[0][0]
__________________________________________________________________________________________________
conv2d_52 (Conv2D) (None, 17, 17, 160) 122880 mixed5[0][0]
__________________________________________________________________________________________________
conv2d_57 (Conv2D) (None, 17, 17, 160) 179200 activation_56[0][0]
__________________________________________________________________________________________________
batch_normalization_52 (BatchNo (None, 17, 17, 160) 480 conv2d_52[0][0]
__________________________________________________________________________________________________
batch_normalization_57 (BatchNo (None, 17, 17, 160) 480 conv2d_57[0][0]
__________________________________________________________________________________________________
activation_52 (Activation) (None, 17, 17, 160) 0 batch_normalization_52[0][0]
__________________________________________________________________________________________________
activation_57 (Activation) (None, 17, 17, 160) 0 batch_normalization_57[0][0]
__________________________________________________________________________________________________
conv2d_53 (Conv2D) (None, 17, 17, 160) 179200 activation_52[0][0]
__________________________________________________________________________________________________
conv2d_58 (Conv2D) (None, 17, 17, 160) 179200 activation_57[0][0]
__________________________________________________________________________________________________
batch_normalization_53 (BatchNo (None, 17, 17, 160) 480 conv2d_53[0][0]
__________________________________________________________________________________________________
batch_normalization_58 (BatchNo (None, 17, 17, 160) 480 conv2d_58[0][0]
__________________________________________________________________________________________________
activation_53 (Activation) (None, 17, 17, 160) 0 batch_normalization_53[0][0]
__________________________________________________________________________________________________
activation_58 (Activation) (None, 17, 17, 160) 0 batch_normalization_58[0][0]
__________________________________________________________________________________________________
average_pooling2d_6 (AveragePoo (None, 17, 17, 768) 0 mixed5[0][0]
__________________________________________________________________________________________________
conv2d_51 (Conv2D) (None, 17, 17, 192) 147456 mixed5[0][0]
__________________________________________________________________________________________________
conv2d_54 (Conv2D) (None, 17, 17, 192) 215040 activation_53[0][0]
__________________________________________________________________________________________________
conv2d_59 (Conv2D) (None, 17, 17, 192) 215040 activation_58[0][0]
__________________________________________________________________________________________________
conv2d_60 (Conv2D) (None, 17, 17, 192) 147456 average_pooling2d_6[0][0]
__________________________________________________________________________________________________
batch_normalization_51 (BatchNo (None, 17, 17, 192) 576 conv2d_51[0][0]
__________________________________________________________________________________________________
batch_normalization_54 (BatchNo (None, 17, 17, 192) 576 conv2d_54[0][0]
__________________________________________________________________________________________________
batch_normalization_59 (BatchNo (None, 17, 17, 192) 576 conv2d_59[0][0]
__________________________________________________________________________________________________
batch_normalization_60 (BatchNo (None, 17, 17, 192) 576 conv2d_60[0][0]
__________________________________________________________________________________________________
activation_51 (Activation) (None, 17, 17, 192) 0 batch_normalization_51[0][0]
__________________________________________________________________________________________________
activation_54 (Activation) (None, 17, 17, 192) 0 batch_normalization_54[0][0]
__________________________________________________________________________________________________
activation_59 (Activation) (None, 17, 17, 192) 0 batch_normalization_59[0][0]
__________________________________________________________________________________________________
activation_60 (Activation) (None, 17, 17, 192) 0 batch_normalization_60[0][0]
__________________________________________________________________________________________________
mixed6 (Concatenate) (None, 17, 17, 768) 0 activation_51[0][0]
activation_54[0][0]
activation_59[0][0]
activation_60[0][0]
__________________________________________________________________________________________________
第197-228层,(17, 17, 768)→【Mixed_6e, InceptionC】→(17, 17, 768)
__________________________________________________________________________________________________
conv2d_65 (Conv2D) (None, 17, 17, 192) 147456 mixed6[0][0]
__________________________________________________________________________________________________
batch_normalization_65 (BatchNo (None, 17, 17, 192) 576 conv2d_65[0][0]
__________________________________________________________________________________________________
activation_65 (Activation) (None, 17, 17, 192) 0 batch_normalization_65[0][0]
__________________________________________________________________________________________________
conv2d_66 (Conv2D) (None, 17, 17, 192) 258048 activation_65[0][0]
__________________________________________________________________________________________________
batch_normalization_66 (BatchNo (None, 17, 17, 192) 576 conv2d_66[0][0]
__________________________________________________________________________________________________
activation_66 (Activation) (None, 17, 17, 192) 0 batch_normalization_66[0][0]
__________________________________________________________________________________________________
conv2d_62 (Conv2D) (None, 17, 17, 192) 147456 mixed6[0][0]
__________________________________________________________________________________________________
conv2d_67 (Conv2D) (None, 17, 17, 192) 258048 activation_66[0][0]
__________________________________________________________________________________________________
batch_normalization_62 (BatchNo (None, 17, 17, 192) 576 conv2d_62[0][0]
__________________________________________________________________________________________________
batch_normalization_67 (BatchNo (None, 17, 17, 192) 576 conv2d_67[0][0]
__________________________________________________________________________________________________
activation_62 (Activation) (None, 17, 17, 192) 0 batch_normalization_62[0][0]
__________________________________________________________________________________________________
activation_67 (Activation) (None, 17, 17, 192) 0 batch_normalization_67[0][0]
__________________________________________________________________________________________________
conv2d_63 (Conv2D) (None, 17, 17, 192) 258048 activation_62[0][0]
__________________________________________________________________________________________________
conv2d_68 (Conv2D) (None, 17, 17, 192) 258048 activation_67[0][0]
__________________________________________________________________________________________________
batch_normalization_63 (BatchNo (None, 17, 17, 192) 576 conv2d_63[0][0]
__________________________________________________________________________________________________
batch_normalization_68 (BatchNo (None, 17, 17, 192) 576 conv2d_68[0][0]
__________________________________________________________________________________________________
activation_63 (Activation) (None, 17, 17, 192) 0 batch_normalization_63[0][0]
__________________________________________________________________________________________________
activation_68 (Activation) (None, 17, 17, 192) 0 batch_normalization_68[0][0]
__________________________________________________________________________________________________
average_pooling2d_7 (AveragePoo (None, 17, 17, 768) 0 mixed6[0][0]
__________________________________________________________________________________________________
conv2d_61 (Conv2D) (None, 17, 17, 192) 147456 mixed6[0][0]
__________________________________________________________________________________________________
conv2d_64 (Conv2D) (None, 17, 17, 192) 258048 activation_63[0][0]
__________________________________________________________________________________________________
conv2d_69 (Conv2D) (None, 17, 17, 192) 258048 activation_68[0][0]
__________________________________________________________________________________________________
conv2d_70 (Conv2D) (None, 17, 17, 192) 147456 average_pooling2d_7[0][0]
__________________________________________________________________________________________________
batch_normalization_61 (BatchNo (None, 17, 17, 192) 576 conv2d_61[0][0]
__________________________________________________________________________________________________
batch_normalization_64 (BatchNo (None, 17, 17, 192) 576 conv2d_64[0][0]
__________________________________________________________________________________________________
batch_normalization_69 (BatchNo (None, 17, 17, 192) 576 conv2d_69[0][0]
__________________________________________________________________________________________________
batch_normalization_70 (BatchNo (None, 17, 17, 192) 576 conv2d_70[0][0]
__________________________________________________________________________________________________
activation_61 (Activation) (None, 17, 17, 192) 0 batch_normalization_61[0][0]
__________________________________________________________________________________________________
activation_64 (Activation) (None, 17, 17, 192) 0 batch_normalization_64[0][0]
__________________________________________________________________________________________________
activation_69 (Activation) (None, 17, 17, 192) 0 batch_normalization_69[0][0]
__________________________________________________________________________________________________
activation_70 (Activation) (None, 17, 17, 192) 0 batch_normalization_70[0][0]
__________________________________________________________________________________________________
mixed7 (Concatenate) (None, 17, 17, 768) 0 activation_61[0][0]
activation_64[0][0]
activation_69[0][0]
activation_70[0][0]
__________________________________________________________________________________________________
第229-248层,(17, 17, 768)→【Mixed_7a, InceptionD】→(8, 8, 1280)
__________________________________________________________________________________________________
conv2d_73 (Conv2D) (None, 17, 17, 192) 147456 mixed7[0][0]
__________________________________________________________________________________________________
batch_normalization_73 (BatchNo (None, 17, 17, 192) 576 conv2d_73[0][0]
__________________________________________________________________________________________________
activation_73 (Activation) (None, 17, 17, 192) 0 batch_normalization_73[0][0]
__________________________________________________________________________________________________
conv2d_74 (Conv2D) (None, 17, 17, 192) 258048 activation_73[0][0]
__________________________________________________________________________________________________
batch_normalization_74 (BatchNo (None, 17, 17, 192) 576 conv2d_74[0][0]
__________________________________________________________________________________________________
activation_74 (Activation) (None, 17, 17, 192) 0 batch_normalization_74[0][0]
__________________________________________________________________________________________________
conv2d_71 (Conv2D) (None, 17, 17, 192) 147456 mixed7[0][0]
__________________________________________________________________________________________________
conv2d_75 (Conv2D) (None, 17, 17, 192) 258048 activation_74[0][0]
__________________________________________________________________________________________________
batch_normalization_71 (BatchNo (None, 17, 17, 192) 576 conv2d_71[0][0]
__________________________________________________________________________________________________
batch_normalization_75 (BatchNo (None, 17, 17, 192) 576 conv2d_75[0][0]
__________________________________________________________________________________________________
activation_71 (Activation) (None, 17, 17, 192) 0 batch_normalization_71[0][0]
__________________________________________________________________________________________________
activation_75 (Activation) (None, 17, 17, 192) 0 batch_normalization_75[0][0]
__________________________________________________________________________________________________
conv2d_72 (Conv2D) (None, 8, 8, 320) 552960 activation_71[0][0]
__________________________________________________________________________________________________
conv2d_76 (Conv2D) (None, 8, 8, 192) 331776 activation_75[0][0]
__________________________________________________________________________________________________
batch_normalization_72 (BatchNo (None, 8, 8, 320) 960 conv2d_72[0][0]
__________________________________________________________________________________________________
batch_normalization_76 (BatchNo (None, 8, 8, 192) 576 conv2d_76[0][0]
__________________________________________________________________________________________________
activation_72 (Activation) (None, 8, 8, 320) 0 batch_normalization_72[0][0]
__________________________________________________________________________________________________
activation_76 (Activation) (None, 8, 8, 192) 0 batch_normalization_76[0][0]
__________________________________________________________________________________________________
max_pooling2d_4 (MaxPooling2D) (None, 8, 8, 768) 0 mixed7[0][0]
__________________________________________________________________________________________________
mixed8 (Concatenate) (None, 8, 8, 1280) 0 activation_72[0][0]
activation_76[0][0]
max_pooling2d_4[0][0]
__________________________________________________________________________________________________
第249-279层,(8, 8, 1280)→【Mixed_7b, InceptionE】→(8, 8, 2048)
__________________________________________________________________________________________________
conv2d_81 (Conv2D) (None, 8, 8, 448) 573440 mixed8[0][0]
__________________________________________________________________________________________________
batch_normalization_81 (BatchNo (None, 8, 8, 448) 1344 conv2d_81[0][0]
__________________________________________________________________________________________________
activation_81 (Activation) (None, 8, 8, 448) 0 batch_normalization_81[0][0]
__________________________________________________________________________________________________
conv2d_78 (Conv2D) (None, 8, 8, 384) 491520 mixed8[0][0]
__________________________________________________________________________________________________
conv2d_82 (Conv2D) (None, 8, 8, 384) 1548288 activation_81[0][0]
__________________________________________________________________________________________________
batch_normalization_78 (BatchNo (None, 8, 8, 384) 1152 conv2d_78[0][0]
__________________________________________________________________________________________________
batch_normalization_82 (BatchNo (None, 8, 8, 384) 1152 conv2d_82[0][0]
__________________________________________________________________________________________________
activation_78 (Activation) (None, 8, 8, 384) 0 batch_normalization_78[0][0]
__________________________________________________________________________________________________
activation_82 (Activation) (None, 8, 8, 384) 0 batch_normalization_82[0][0]
__________________________________________________________________________________________________
conv2d_79 (Conv2D) (None, 8, 8, 384) 442368 activation_78[0][0]
__________________________________________________________________________________________________
conv2d_80 (Conv2D) (None, 8, 8, 384) 442368 activation_78[0][0]
__________________________________________________________________________________________________
conv2d_83 (Conv2D) (None, 8, 8, 384) 442368 activation_82[0][0]
__________________________________________________________________________________________________
conv2d_84 (Conv2D) (None, 8, 8, 384) 442368 activation_82[0][0]
__________________________________________________________________________________________________
average_pooling2d_8 (AveragePoo (None, 8, 8, 1280) 0 mixed8[0][0]
__________________________________________________________________________________________________
conv2d_77 (Conv2D) (None, 8, 8, 320) 409600 mixed8[0][0]
__________________________________________________________________________________________________
batch_normalization_79 (BatchNo (None, 8, 8, 384) 1152 conv2d_79[0][0]
__________________________________________________________________________________________________
batch_normalization_80 (BatchNo (None, 8, 8, 384) 1152 conv2d_80[0][0]
__________________________________________________________________________________________________
batch_normalization_83 (BatchNo (None, 8, 8, 384) 1152 conv2d_83[0][0]
__________________________________________________________________________________________________
batch_normalization_84 (BatchNo (None, 8, 8, 384) 1152 conv2d_84[0][0]
__________________________________________________________________________________________________
conv2d_85 (Conv2D) (None, 8, 8, 192) 245760 average_pooling2d_8[0][0]
__________________________________________________________________________________________________
batch_normalization_77 (BatchNo (None, 8, 8, 320) 960 conv2d_77[0][0]
__________________________________________________________________________________________________
activation_79 (Activation) (None, 8, 8, 384) 0 batch_normalization_79[0][0]
__________________________________________________________________________________________________
activation_80 (Activation) (None, 8, 8, 384) 0 batch_normalization_80[0][0]
__________________________________________________________________________________________________
activation_83 (Activation) (None, 8, 8, 384) 0 batch_normalization_83[0][0]
__________________________________________________________________________________________________
activation_84 (Activation) (None, 8, 8, 384) 0 batch_normalization_84[0][0]
__________________________________________________________________________________________________
batch_normalization_85 (BatchNo (None, 8, 8, 192) 576 conv2d_85[0][0]
__________________________________________________________________________________________________
activation_77 (Activation) (None, 8, 8, 320) 0 batch_normalization_77[0][0]
__________________________________________________________________________________________________
mixed9_0 (Concatenate) (None, 8, 8, 768) 0 activation_79[0][0]
activation_80[0][0]
__________________________________________________________________________________________________
concatenate_1 (Concatenate) (None, 8, 8, 768) 0 activation_83[0][0]
activation_84[0][0]
__________________________________________________________________________________________________
activation_85 (Activation) (None, 8, 8, 192) 0 batch_normalization_85[0][0]
__________________________________________________________________________________________________
mixed9 (Concatenate) (None, 8, 8, 2048) 0 activation_77[0][0]
mixed9_0[0][0]
concatenate_1[0][0]
activation_85[0][0]
__________________________________________________________________________________________________
第280-310层,(8, 8, 2048)→【Mixed_7c, InceptionE】→(8, 8, 2048)
__________________________________________________________________________________________________
conv2d_90 (Conv2D) (None, 8, 8, 448) 917504 mixed9[0][0]
__________________________________________________________________________________________________
batch_normalization_90 (BatchNo (None, 8, 8, 448) 1344 conv2d_90[0][0]
__________________________________________________________________________________________________
activation_90 (Activation) (None, 8, 8, 448) 0 batch_normalization_90[0][0]
__________________________________________________________________________________________________
conv2d_87 (Conv2D) (None, 8, 8, 384) 786432 mixed9[0][0]
__________________________________________________________________________________________________
conv2d_91 (Conv2D) (None, 8, 8, 384) 1548288 activation_90[0][0]
__________________________________________________________________________________________________
batch_normalization_87 (BatchNo (None, 8, 8, 384) 1152 conv2d_87[0][0]
__________________________________________________________________________________________________
batch_normalization_91 (BatchNo (None, 8, 8, 384) 1152 conv2d_91[0][0]
__________________________________________________________________________________________________
activation_87 (Activation) (None, 8, 8, 384) 0 batch_normalization_87[0][0]
__________________________________________________________________________________________________
activation_91 (Activation) (None, 8, 8, 384) 0 batch_normalization_91[0][0]
__________________________________________________________________________________________________
conv2d_88 (Conv2D) (None, 8, 8, 384) 442368 activation_87[0][0]
__________________________________________________________________________________________________
conv2d_89 (Conv2D) (None, 8, 8, 384) 442368 activation_87[0][0]
__________________________________________________________________________________________________
conv2d_92 (Conv2D) (None, 8, 8, 384) 442368 activation_91[0][0]
__________________________________________________________________________________________________
conv2d_93 (Conv2D) (None, 8, 8, 384) 442368 activation_91[0][0]
__________________________________________________________________________________________________
average_pooling2d_9 (AveragePoo (None, 8, 8, 2048) 0 mixed9[0][0]
__________________________________________________________________________________________________
conv2d_86 (Conv2D) (None, 8, 8, 320) 655360 mixed9[0][0]
__________________________________________________________________________________________________
batch_normalization_88 (BatchNo (None, 8, 8, 384) 1152 conv2d_88[0][0]
__________________________________________________________________________________________________
batch_normalization_89 (BatchNo (None, 8, 8, 384) 1152 conv2d_89[0][0]
__________________________________________________________________________________________________
batch_normalization_92 (BatchNo (None, 8, 8, 384) 1152 conv2d_92[0][0]
__________________________________________________________________________________________________
batch_normalization_93 (BatchNo (None, 8, 8, 384) 1152 conv2d_93[0][0]
__________________________________________________________________________________________________
conv2d_94 (Conv2D) (None, 8, 8, 192) 393216 average_pooling2d_9[0][0]
__________________________________________________________________________________________________
batch_normalization_86 (BatchNo (None, 8, 8, 320) 960 conv2d_86[0][0]
__________________________________________________________________________________________________
activation_88 (Activation) (None, 8, 8, 384) 0 batch_normalization_88[0][0]
__________________________________________________________________________________________________
activation_89 (Activation) (None, 8, 8, 384) 0 batch_normalization_89[0][0]
__________________________________________________________________________________________________
activation_92 (Activation) (None, 8, 8, 384) 0 batch_normalization_92[0][0]
__________________________________________________________________________________________________
activation_93 (Activation) (None, 8, 8, 384) 0 batch_normalization_93[0][0]
__________________________________________________________________________________________________
batch_normalization_94 (BatchNo (None, 8, 8, 192) 576 conv2d_94[0][0]
__________________________________________________________________________________________________
activation_86 (Activation) (None, 8, 8, 320) 0 batch_normalization_86[0][0]
__________________________________________________________________________________________________
mixed9_1 (Concatenate) (None, 8, 8, 768) 0 activation_88[0][0]
activation_89[0][0]
__________________________________________________________________________________________________
concatenate_2 (Concatenate) (None, 8, 8, 768) 0 activation_92[0][0]
activation_93[0][0]
__________________________________________________________________________________________________
activation_94 (Activation) (None, 8, 8, 192) 0 batch_normalization_94[0][0]
__________________________________________________________________________________________________
mixed10 (Concatenate) (None, 8, 8, 2048) 0 activation_86[0][0]
mixed9_1[0][0]
concatenate_2[0][0]
activation_94[0][0]
__________________________________________________________________________________________________
第311-312层,(8, 8, 2048)→【global avg pool】→(2048,)→【dropout】→(2048,)→【fc】→(1000,)
__________________________________________________________________________________________________
avg_pool (GlobalAveragePooling2 (None, 2048) 0 mixed10[0][0]
__________________________________________________________________________________________________
predictions (Dense) (None, 1000) 2049000 avg_pool[0][0]
==================================================================================================
Total params: 23,851,784
Trainable params: 23,817,352
Non-trainable params: 34,432
__________________________________________________________________________________________________
Non-trainable params参数均来自bn层
源代码中,x = layers.BatchNormalization(axis=bn_axis, scale=False, name=bn_name)(x)
,设置scale=False
表示省略参数
γ
\gamma
γ(源代码中对应gamma
),理由是:When the next layer is linear (also e.g. nn.relu), this can be disabled since the scaling will be done by the next layer.
例如第1个bn层,(149, 149, 32)→【bn】→(149, 149, 32)
,分别对32个通道进行batch norm,每个通道包含3个参数,即
μ
,
σ
,
β
\mu, \sigma, \beta
μ,σ,β(源代码中分别对应moving_mean, moving_variance, beta
),其中
μ
,
σ
\mu, \sigma
μ,σ是Non-trainable的参数,
β
\beta
β是trainable的参数
故该bn层参数总数为:32×3=96,Non-trainable的参数总数为:32×2=64