name: "AlexNet"
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param { # 对输入做227*227的随机裁剪，同时做镜像来扩大样本数量，来降低过拟合的问题。按照alex论文的说法，TRAIN会扩大2048倍的样本裁剪，TEST会生成10个新样本，四个角和居中裁剪，以及镜像（但是这个caffe实现test不做镜像）
mirror: true
crop_size: 227
mean_file: "data/ilsvrc12/imagenet_mean.binaryproto" #对图片做零均值化。这个我目前的理解是一种归一化的手段。训练图片有的颜色浓，有的颜色淡，通过该方法，可以将每张照片的数据分布基于（0，0）坐标原点来分布，可以标准化训练样本。参见https://my.oschina.net/findbill/blog/661817
}
data_param {
source: "examples/imagenet/ilsvrc12_train_lmdb"
batch_size: 256 #一次处理的图片数量
backend: LMDB
}
}
layer {
name: "data"
type: "Data"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
mirror: false
crop_size: 227
mean_file: "data/ilsvrc12/imagenet_mean.binaryproto"
}
data_param {
source: "examples/imagenet/ilsvrc12_val_lmdb"
batch_size: 50
backend: LMDB
}
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
param {# weigith的学习速率设置
lr_mult: 1 # 学习速率系数，这个乘以solver.prototxt 配置文件中的 base_lr，就是这一层的初始学习率
decay_mult: 1 #衰减系数，避免过拟合，但是细节原理还不懂TODO
}
param {# bias的学习速率设置
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 96 #多少个卷积核，也就说输入图片通过卷积生成多少个特征。
kernel_size: 11 # 卷积核的大小
stride: 4 #卷积滑动的步长
weight_filler { #权重初始化
type: "gaussian" # 权重初始化使用高斯分布
std: 0.01 #标准差为0.01， 均值默认为0
}
bias_filler { #偏置初始化
type: "constant"
value: 0
}
}
}
layer {
name: "relu1"
type: "ReLU" # 过滤掉CONV1输出<0的输出，这个我从其他文章看过来自己的理解是因为relu更接近人类神经元的激活函数。人类的神经元的对一个输入的激活只有5%，通过relu可以降低神经元的激活数量
bottom: "conv1"
top: "conv1"
}
layer {
name: "norm1"
type: "LRN" #局部归一化，还不懂TODO
bottom: "conv1"
top: "norm1"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool1"
type: "Pooling" #池化，用来降低位置相关性，我认为也是讲特征进一步标准化的手段。
bottom: "norm1"
top: "pool1"
pooling_param {
pool: MAX # max(0,x)
kernel_size: 3 #核的尺寸
stride: 2 #步长，这个步长形成了一个重叠滑动的池化动作，alex论文讲这样比不做重叠可以稍微改善过拟合的情况
}
}
layer {
name: "conv2"
type: "Convolution"
bottom: "pool1"
top: "conv2"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 2 # 给输入四个边都加上2个空白像素
kernel_size: 5
group: 2 # 将输入卷积运算分成两个组运算。这个是解决GPU内存不足来用的
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "norm2"
type: "LRN"
bottom: "conv2"
top: "norm2"
lrn_param {
local_size: 5
alpha: 0.0001
beta: 0.75
}
}
layer {
name: "pool2"
type: "Pooling"
bottom: "norm2"
top: "pool2"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "conv3"
type: "Convolution"
bottom: "pool2"
top: "conv3"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv3"
top: "conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 384
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv4"
top: "conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 256
pad: 1
kernel_size: 3
group: 2
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "pool5"
type: "Pooling"
bottom: "conv5"
top: "pool5"
pooling_param {
pool: MAX
kernel_size: 3
stride: 2
}
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "pool5"
top: "fc6"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu6"
type: "ReLU"
bottom: "fc6"
top: "fc6"
}
layer {
name: "drop6"
type: "Dropout" # 丢掉一定数量的神经元，不做输出，也不参与反向传播的权值计算。这个的目的是降低神经元之间的固定依赖性。每次迭代神经元的链接通道都是随机的，从而避免固定依赖。
bottom: "fc6"
top: "fc6"
dropout_param {
dropout_ratio: 0.5 # 丢弃50%的神经元
}
}
layer {
name: "fc7"
type: "InnerProduct"
bottom: "fc6"
top: "fc7"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 4096
weight_filler {
type: "gaussian"
std: 0.005
}
bias_filler {
type: "constant"
value: 0.1
}
}
}
layer {
name: "relu7"
type: "ReLU"
bottom: "fc7"
top: "fc7"
}
layer {
name: "drop7"
type: "Dropout"
bottom: "fc7"
top: "fc7"
dropout_param {
dropout_ratio: 0.5
}
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "fc7"
top: "fc8"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
inner_product_param {
num_output: 1000
weight_filler {
type: "gaussian"
std: 0.01
}
bias_filler {
type: "constant"
value: 0
}
}
}
layer {
name: "accuracy"
type: "Accuracy"
bottom: "fc8"
bottom: "label"
top: "accuracy"
include {
phase: TEST
}
}
layer {
name: "loss"
type: "SoftmaxWithLoss"
bottom: "fc8"
bottom: "label"
top: "loss"
}
