根據(jù) *_train_test.prototxt文件生成 *_deploy.prototxt文件

發(fā)表于2016/8/6 19:43:11 ?1218人閱讀

本文參考博文

（1）介紹?*_train_test.prototxt文件與 *_deploy.prototxt文件的不同：http://blog.csdn.net/sunshine_in_moon/article/details/49472901????

（2）生成deploy文件的Python代碼：http://www.cnblogs.com/denny402/p/5685818.html???????

*_train_test.prototxt文件

這是訓(xùn)練與測試網(wǎng)絡(luò)配置文件

*_deploy.prototxt文件 這是模型構(gòu)造文件

在博文http://www.cnblogs.com/denny402/p/5685818.html?????中給出了生成 deploy.prototxt文件的Python源代碼，但是每個網(wǎng)絡(luò)不同，修改起來比較麻煩，下面給出該博文中以mnist為例生成deploy文件的源代碼，可根據(jù)自己網(wǎng)絡(luò)的設(shè)置做出相應(yīng)修改：（下方代碼未測試）

# -*- coding: utf-8 -*-from caffe import layers as L,params as P,to_proto root='/home/xxx/' deploy=root+'mnist/deploy.prototxt' #文件保存路徑def create_deploy():#少了第一層，data層conv1=L.Convolution(bottom='data', kernel_size=5, stride=1,num_output=20, pad=0,weight_filler=dict(type='xavier'))pool1=L.Pooling(conv1, pool=P.Pooling.MAX, kernel_size=2, stride=2)conv2=L.Convolution(pool1, kernel_size=5, stride=1,num_output=50, pad=0,weight_filler=dict(type='xavier'))pool2=L.Pooling(conv2, pool=P.Pooling.MAX, kernel_size=2, stride=2)fc3=L.InnerProduct(pool2, num_output=500,weight_filler=dict(type='xavier'))relu3=L.ReLU(fc3, in_place=True)fc4 = L.InnerProduct(relu3, num_output=10,weight_filler=dict(type='xavier'))#最后沒有accuracy層，但有一個Softmax層prob=L.Softmax(fc4)return to_proto(prob) def write_deploy(): with open(deploy, 'w') as f:f.write('name:"Lenet"\n')f.write('input:"data"\n')f.write('input_dim:1\n')f.write('input_dim:3\n')f.write('input_dim:28\n')f.write('input_dim:28\n')f.write(str(create_deploy())) if __name__ == '__main__':write_deploy()

用代碼生成deploy文件還是比較麻煩。我們在構(gòu)建深度學(xué)習(xí)網(wǎng)絡(luò)時，肯定會先定義好訓(xùn)練與測試網(wǎng)絡(luò)的配置文件——*_train_test.prototxt文件，我們可以通過修改*_train_test.prototxt文件 來生成 deploy 文件。以cifar10為例先簡單介紹一下兩者的區(qū)別。

（1）deploy 文件中的數(shù)據(jù)層更為簡單，即將*_train_test.prototxt文件中的輸入訓(xùn)練數(shù)據(jù)lmdb與輸入測試數(shù)據(jù)lmdb這兩層刪除，取而代之的是，

layer {name: "data"type: "Input"top: "data"input_param { shape: { dim: 1 dim: 3 dim: 32 dim: 32 } } }

注：shape: { dim: 1 dim: 3 dim: 32 dim: 32 }代表含義：
shape {dim: 1 #num，對待識別樣本進(jìn)行數(shù)據(jù)增廣的數(shù)量，可自行定義。一般會進(jìn)行5次crop，之后分別flip。如果該值為10則表示一個樣本會變成10個，之后輸入到網(wǎng)絡(luò)進(jìn)行識別。如果不進(jìn)行數(shù)據(jù)增廣，可以設(shè)置成1dim: 3 #通道數(shù)，表示RGB三個通道dim: 32 #圖像的長和寬，通過 *_train_test.prototxt文件中數(shù)據(jù)輸入層的crop_size獲取dim: 32

（2）卷積層和全連接層中weight_filler{}與bias_filler{}兩個參數(shù)不用再填寫，因為這兩個參數(shù)的值，由已經(jīng)訓(xùn)練好的模型*.caffemodel文件提供。如下所示代碼，將*_train_test.prototxt文件中的weight_filler、bias_filler全部刪除。

layer {????????????????????????????? # weight_filler、bias_filler刪除
? name: "ip2"
? type: "InnerProduct"
? bottom: "ip1"
? top: "ip2"
? param {
??? lr_mult: 1?? #權(quán)重w的學(xué)習(xí)率倍數(shù)
? }
? param {
??? lr_mult: 2??? #偏置b的學(xué)習(xí)率倍數(shù)
? }
? inner_product_param {
??? num_output: 10
??? weight_filler {
????? type: "gaussian"
????? std: 0.1
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}

刪除后變?yōu)?/p>

layer { name: "ip2"type: "InnerProduct"bottom: "ip1"top: "ip2"param {lr_mult: 1}param {lr_mult: 2}inner_product_param {num_output: 10} }
（3）輸出層的變化?? ???? 1）沒有了test模塊測試精度?，將該層刪除????? ?????2）輸出層
1）*_deploy.prototxt文件的構(gòu)造和*_train_test.prototxt文件的構(gòu)造最為明顯的不同點是，deploy文件沒有test網(wǎng)絡(luò)中的test模塊，只有訓(xùn)練模塊，即將*_train_test.prototxt中最后部分的test模塊測試精度刪除，即將如下代碼刪除。
layer { #刪除該層name: "accuracy"type: "Accuracy"bottom: "ip2"bottom: "label"top: "accuracy"include {phase: TEST} }

2） 輸出層?

*_train_test.prototxt文件

layer{name: "loss" #注意此處層名稱與下面的不同type: "SoftmaxWithLoss" #注意此處與下面的不同bottom: "ip2"bottom: "label" #注意標(biāo)簽項在下面沒有了，因為下面的預(yù)測屬于哪個標(biāo)簽，因此不能提供標(biāo)簽top: "loss" }

*_deploy.prototxt文件

layer {name: "prob"type: "Softmax"bottom: "ip2"top: "prob" }

注意在兩個文件中輸出層的類型都發(fā)生了變化一個是SoftmaxWithLoss，另一個是Softmax。另外為了方便區(qū)分訓(xùn)練與應(yīng)用輸出，訓(xùn)練是輸出時是loss，應(yīng)用時是prob。

下面給出CIFAR10中的配置文件cifar10_quick_train_test.prototxt與其模型構(gòu)造文件? cifar10_quick.prototxt 直觀展示兩者的區(qū)別。

cifar10_quick_train_test.prototxt文件代碼

name: "CIFAR10_quick"
layer {?????????????? #該層去掉
? name: "cifar"
? type: "Data"
? top: "data"
? top: "label"
? include {
??? phase: TRAIN
? }
? transform_param {
??? mean_file: "examples/cifar10/mean.binaryproto"
? }
? data_param {
??? source: "examples/cifar10/cifar10_train_lmdb"
??? batch_size: 100
??? backend: LMDB
? }
}
layer {???????????? #該層去掉
? name: "cifar"
? type: "Data"
? top: "data"
? top: "label"
? include {
??? phase: TEST
? }
? transform_param {
??? mean_file: "examples/cifar10/mean.binaryproto"
? }
? data_param {
??? source: "examples/cifar10/cifar10_test_lmdb"
??? batch_size: 100
??? backend: LMDB
? }
}
layer {??????????????????????? #將下方的weight_filler、bias_filler全部刪除
? name: "conv1"
? type: "Convolution"
? bottom: "data"
? top: "conv1"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? convolution_param {
??? num_output: 32
??? pad: 2
??? kernel_size: 5
??? stride: 1
??? weight_filler {
????? type: "gaussian"
????? std: 0.0001
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}
layer {
? name: "pool1"
? type: "Pooling"
? bottom: "conv1"
? top: "pool1"
? pooling_param {
??? pool: MAX
??? kernel_size: 3
??? stride: 2
? }
}
layer {
? name: "relu1"
? type: "ReLU"
? bottom: "pool1"
? top: "pool1"
}
layer {???????????????????????? #weight_filler、bias_filler刪除
? name: "conv2"
? type: "Convolution"
? bottom: "pool1"
? top: "conv2"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? convolution_param {
??? num_output: 32
??? pad: 2
??? kernel_size: 5
??? stride: 1
??? weight_filler {
????? type: "gaussian"
????? std: 0.01
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}
layer {
? name: "relu2"
? type: "ReLU"
? bottom: "conv2"
? top: "conv2"
}
layer {
? name: "pool2"
? type: "Pooling"
? bottom: "conv2"
? top: "pool2"
? pooling_param {
??? pool: AVE
??? kernel_size: 3
??? stride: 2
? }
}
layer {???????????????????????? #weight_filler、bias_filler刪除
? name: "conv3"
? type: "Convolution"
? bottom: "pool2"
? top: "conv3"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? convolution_param {
??? num_output: 64
??? pad: 2
??? kernel_size: 5
??? stride: 1
??? weight_filler {
????? type: "gaussian"
????? std: 0.01
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}
layer {
? name: "relu3"
? type: "ReLU"
? bottom: "conv3"
? top: "conv3"
}
layer {
? name: "pool3"
? type: "Pooling"
? bottom: "conv3"
? top: "pool3"
? pooling_param {
??? pool: AVE
??? kernel_size: 3
??? stride: 2
? }
}
layer {?????????????????????? #weight_filler、bias_filler刪除
? name: "ip1"
? type: "InnerProduct"
? bottom: "pool3"
? top: "ip1"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? inner_product_param {
??? num_output: 64
??? weight_filler {
????? type: "gaussian"
????? std: 0.1
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}
layer {????????????????????????????? # weight_filler、bias_filler刪除
? name: "ip2"
? type: "InnerProduct"
? bottom: "ip1"
? top: "ip2"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? inner_product_param {
??? num_output: 10
??? weight_filler {
????? type: "gaussian"
????? std: 0.1
??? }
??? bias_filler {
????? type: "constant"
??? }
? }
}
layer {????????????????????????????????? #將該層刪除
? name: "accuracy"
? type: "Accuracy"
? bottom: "ip2"
? bottom: "label"
? top: "accuracy"
? include {
??? phase: TEST
? }
}
layer {???????????????????????????????? #修改
? name: "loss"?????? #---loss? 修改為? prob
? type: "SoftmaxWithLoss"??????????? ?# SoftmaxWithLoss 修改為 softmax
? bottom: "ip2"
? bottom: "label"????????? #去掉
? top: "loss"
}

以下為cifar10_quick.prototxt

layer {?????????????? #將兩個輸入層修改為該層
? name: "data"
? type: "Input"
? top: "data"
? input_param { shape: { dim: 1 dim: 3 dim: 32 dim: 32 } }???? #注意shape中變量值的修改，CIFAR10中的 *_train_test.protxt文件中沒有 crop_size
}
layer {
? name: "conv1"
? type: "Convolution"
? bottom: "data"
? top: "conv1"
? param {
??? lr_mult: 1?? #權(quán)重W的學(xué)習(xí)率倍數(shù)
}
? param {
??? lr_mult: 2?? #偏置b的學(xué)習(xí)率倍數(shù)
? }
? convolution_param {
??? num_output: 32
??? pad: 2???#加邊為2
?? kernel_size: 5
??? stride: 1
? }
}
layer {
? name: "pool1"
? type: "Pooling"
? bottom: "conv1"
? top: "pool1"
? pooling_param {
??? pool: MAX????#Max Pooling
?? kernel_size: 3
??? stride: 2
? }
}
layer {
? name: "relu1"
? type: "ReLU"
? bottom: "pool1"
? top: "pool1"
}
layer {
? name: "conv2"
? type: "Convolution"
? bottom: "pool1"
? top: "conv2"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? convolution_param {
??? num_output: 32
??? pad: 2
??? kernel_size: 5
??? stride: 1
? }
}
layer {
? name: "relu2"
? type: "ReLU"
? bottom: "conv2"
? top: "conv2"
}
layer {
? name: "pool2"
? type: "Pooling"
? bottom: "conv2"
? top: "pool2"
? pooling_param {
??? pool: AVE?? #均值池化
??? kernel_size: 3
??? stride: 2
? }
}
layer {
? name: "conv3"
? type: "Convolution"
? bottom: "pool2"
? top: "conv3"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? convolution_param {
??? num_output: 64
??? pad: 2
??? kernel_size: 5
??? stride: 1
? }
}
layer {
? name: "relu3"
? type: "ReLU"? #使用ReLU激勵函數(shù)，這里需要注意的是，本層的bottom和top都是conv3>
? bottom: "conv3"
? top: "conv3"
}
layer {
? name: "pool3"
? type: "Pooling"
? bottom: "conv3"
? top: "pool3"
? pooling_param {
??? pool: AVE
kernel_size: 3
??? stride: 2
? }
}
layer {
? name: "ip1"
? type: "InnerProduct"
? bottom: "pool3"
? top: "ip1"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? inner_product_param {
??? num_output: 64
? }
}
layer {
? name: "ip2"
? type: "InnerProduct"
? bottom: "ip1"
? top: "ip2"
? param {
??? lr_mult: 1
? }
? param {
??? lr_mult: 2
? }
? inner_product_param {
??? num_output: 10
? }
}
layer {
? name: "prob"
? type: "Softmax"
? bottom: "ip2"
? top: "prob"
}

總結(jié)

以上是生活随笔為你收集整理的根据 *_train_test.prototxt文件生成 *_deploy.prototxt文件的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。