本文的目的是熟悉inception v1網(wǎng)絡(luò)結(jié)構(gòu)亏推，以便對(duì)tvm前端有更深入的了解术裸。
網(wǎng)絡(luò)結(jié)構(gòu)可以參考TensorFlow實(shí)現(xiàn)Inception系列結(jié)構(gòu)這篇文章中的圖缰猴。
論文在此：
Going Deeper with Convolutions
http://arxiv.org/abs/1409.4842
查看caffe2的inception v1模型文件，只考慮推理softmax2医男，其中包含的op有：
count operation
1 type: "AveragePool"
1 type: "Dropout"
1 type: "FC"
1 type: "Softmax"
2 type: "LRN"
9 type: "Concat"
13 type: "MaxPool"
57 type: "Conv"
57 type: "Relu"
其中，LRN就是LocalRespNorm捻勉，遺留自AlexNet镀梭，后面會(huì)被BN層代替。Dropout在最后的FC前面踱启。
keras代碼實(shí)現(xiàn)：
caffe2的inception v2模型文件
論文： Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift, 4.8% test error
http://arxiv.org/abs/1502.03167
v2對(duì)v1的改進(jìn)：
- 加入了BN層
- 把5x5的卷積改成了2個(gè)串行的3x3卷積
cout operation
1 type: "FC"
1 type: "Softmax"
5 type: "MaxPool"
8 type: "AveragePool"
10 type: "Concat"
69 type: "Add"
69 type: "Conv"
69 type: "Mul"
69 type: "Relu"
69 type: "SpatialBN"
因此报账，從op上看：多了ADD，MUL埠偿，SpatialBN透罢，少了LRN，DropOut冠蒋。
inception v2 tensorflow源碼實(shí)現(xiàn)
inception v3論文：
Rethinking the Inception Architecture for Computer Vision, 3.5% test error
http://arxiv.org/abs/1512.00567
caffe2模型文件羽圃，非官方提供，從caffe轉(zhuǎn)換而來(lái)：
1 type: "Dropout"
1 type: "FC"
1 type: "Softmax"
5 type: "MaxPool"
9 type: "AveragePool"
11 type: "Concat"
94 type: "Add"
94 type: "Conv"
94 type: "Mul"
94 type: "Relu"
94 type: "SpatialBN"
從op上看抖剿，和V2沒(méi)什么變化朽寞。本身的改進(jìn)也主要在卷積核的大小從nxn變成了1xn和nx1。
inception v4論文：
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning, 3.08% test error
http://arxiv.org/abs/1602.07261
inception v4 和 inception - resnet v2代碼實(shí)現(xiàn)
https://github.com/titu1994/Inception-v4
https://github.com/tensorflow/models/blob/master/research/slim/nets/inception_resnet_v2.py
https://github.com/tensorflow/models/blob/master/research/slim/nets/inception_v4.py
mxnet源碼實(shí)現(xiàn)
https://github.com/apache/incubator-mxnet/blob/master/example/image-classification/symbols/inception-v4.py
沒(méi)找到caffe2的prototxt文件斩郎，只有caffe模型文件 ：
count operation
1 type: "constant"
1 type: "xavier"
1 type: "Dropout"
1 type: "InnerProduct"
1 type: "Softmax"
19 type: "Concat"
19 type: "Pooling"
149 type: "BatchNorm"
149 type: "Convolution"
149 type: "ReLU"
149 type: "Scale"
關(guān)于 inception系列模型更詳細(xì)的資料參考：
https://blog.csdn.net/hejin_some/article/details/78636586
