【一】以 lenet_train_test.prototxt 为例

name: "LeNet"
layer {name: "mnist"type: "Data"top: "data"top: "label"include {phase: TRAIN}transform_param {scale: 0.00390625}data_param {source: "examples/mnist/mnist_train_lmdb"batch_size: 64backend: LMDB}
}
layer {name: "mnist"type: "Data"top: "data"top: "label"include {phase: TEST}transform_param {scale: 0.00390625}data_param {source: "examples/mnist/mnist_test_lmdb"batch_size: 100backend: LMDB}
}
layer {name: "conv1"type: "Convolution"bottom: "data"top: "conv1"param {lr_mult: 1}param {lr_mult: 2}convolution_param {num_output: 20kernel_size: 5stride: 1weight_filler {type: "xavier"}bias_filler {type: "constant"}}
}
layer {name: "pool1"type: "Pooling"bottom: "conv1"top: "pool1"pooling_param {pool: MAXkernel_size: 2stride: 2}
}
layer {name: "conv2"type: "Convolution"bottom: "pool1"top: "conv2"param {lr_mult: 1}param {lr_mult: 2}convolution_param {num_output: 50kernel_size: 5stride: 1weight_filler {type: "xavier"}bias_filler {type: "constant"}}
}
layer {name: "pool2"type: "Pooling"bottom: "conv2"top: "pool2"pooling_param {pool: MAXkernel_size: 2stride: 2}
}
layer {name: "ip1"type: "InnerProduct"bottom: "pool2"top: "ip1"param {lr_mult: 1}param {lr_mult: 2}inner_product_param {num_output: 500weight_filler {type: "xavier"}bias_filler {type: "constant"}}
}
layer {name: "relu1"type: "ReLU"bottom: "ip1"top: "ip1"
}
layer {name: "ip2"type: "InnerProduct"bottom: "ip1"top: "ip2"param {lr_mult: 1}param {lr_mult: 2}inner_product_param {num_output: 10weight_filler {type: "xavier"}bias_filler {type: "constant"}}
}
layer {name: "accuracy"type: "Accuracy"bottom: "ip2"bottom: "label"top: "accuracy"include {phase: TEST}
}
layer {name: "loss"type: "SoftmaxWithLoss"bottom: "ip2"bottom: "label"top: "loss"
}
}

【二】参数详解

包级参数

name ：定义网络名称

可以随便起，例子中为"LeNet"

layer {定义层结构

name ：层的名称

例如，输入层名称为“mnist”

type ：层的类型

例如，输入层的类型为“Data”、卷积层“Convolution”、池化层“Pooling”、全连接层“InnerProduct”、激活层“ReLU”、计算精度“Accuracy”、Softmax损失层“SoftmaxWithLoss”

top ：用来标记输出数据（blob）。例如，在训练中，输入层有两个top分别为“data”和“label”，分别存放了数据和标签。

bottom ：用来标记输入数据（blob）

它的上层的top相同，输入层是顶层，因此没有bottom blob。如果该层的bottom和top相同，则该层的输入和输出占用了一个blob。

include {当前层所属的阶段

一般训练的时候和测试的时候，模型的层是不一样的。该层（layer）是属于训练阶段的层，还是属于测试阶段的层，需要用include来指定。 如果没有include参数，则表示该层既在训练模型中，又在测试模型中。

 phase：当前层的阶段；TRAIN：训练；TEST：测试；
}

data_param {定义数据参数、数据来源等等

source		: 包含数据库的目录名称，如examples/mnist/mnist_train_lmdb
batch_size	: 每次处理的数据个数，如64
rand_skip		: 在开始的时候，路过某个数据的输入。通常对异步的SGD很有用。
backend		: 选择是采用LevelDB还是LMDB, 默认是LevelDB.
}

transform_param {数据的预处理，可以将数据变换到定义的范围内。

scale ：0.00390625，实际上就是1/255, 即将输入数据由0-255归一化到0-1之间
mean_file_size: "examples/cifar10/mean.binaryproto" ，用一个配置文件来进行均值操作
mirror: 1  ，1表示开启镜像，0表示关闭，也可用ture和false来表示
crop_size: 227	 ，剪裁一个 227*227的图块，在训练阶段随机剪裁，在测试阶段从中间裁剪
}

param {定义weight或bias的学习速率和衰减因子参数。

  lr_mult 		： weight或bias的学习速率decay_mult 	：weight或bias的衰减因子}

convolution_param {定义卷积层的参数。

  num_output 	：输出的个数；即卷积核的个数；pad 			：填充边缘的大小。可使用pad_h和pad_w来分别设定kernel_size 	：卷积核的大小，如果卷积核的长和宽不等，需要用kernel_h和kernel_w分别设定group 		：分组，默认为1组。卷积分组可以减少网络的参数。 weight_filler ：权值初始化，若设置为constant, 则默认为0。也可使用"xavier"或者”gaussian"进行初始化bias_filler 	：偏置项的初始化，参考weight_filler
}

lrn_param {定义归一化层的参数

  local_size 	：对于cross channel LRN为需要求和的邻近channel的数量；对于within channel LRN为需要求和的空间区域的边长alpha 		：缩放参数；默认为1；beta 			：β系数；默认为5norm_region 	： [default ACROSS_CHANNELS]: 选择哪种LRN的方法ACROSS_CHANNELS 或者WITHIN_CHANNEL
}

pooling_param {定义池化层的参数

  kernel_size 	：必选；(or kernel_h and kernel_w)：过滤器的大小pool 			：可选；池化的方法，最大值池化MAX、平均值池化AVE 和 随机池化STOCHASTIC；默认MAXstride 		：可选(or stride_h and stride_w) [default1]：指定过滤器的步长
}

inner_product_param {定义全连接层的参数

inner_product_param级别参数num_output 	：过滤器个数。必选；weight_filler	：[default type: 'constant' value: 0]：参数的初始化方法bias_filler 	：偏置的初始化方法 {weight_filler 和 bias_filler级别参数type 	：初始化类型；有 常量“constant” 高斯分布“gaussian” “uniform” “Xavier”等等。常用的是“constant”和“gaussian”std 	：标准差；如果是 gaussian，则可选的设置它；默认值为1；mean 	：均值；如果是 gaussian，则可选的设置它；默认值为0；value 	：定值；如果是“constant”，则设置它；默认值为0；}
}

dropout_param ：定义Dropout层的参数

accuracy_param ：定义Accuracy层的参数

}
}

【三】官方说明

参见：caffe/src/caffe/proto/caffe.proto中message ParamSpec

// Specifies training parameters (multipliers on global learning constants, and the name and other settings used for weight sharing).
// 指定训练参数(全局学习常数的乘数，以及用于权重共享的名称和其他设置)。
message ParamSpec {// The names of the parameter blobs -- useful for sharing parameters among layers, but never required otherwise.  To share a parameter between two layers, give it a (non-empty) name.// 参数块的名称——用于在层之间共享参数，但在其他情况下不需要。要在两个层之间共享一个参数，请给它一个(非空)名称。optional string name = 1;// Whether to require shared weights to have the same shape, or just the same count -- defaults to STRICT if unspecified.// 是否要求共享的权重具有相同的形状，或者只是相同的计数——如果未指定，默认为STRICT。optional DimCheckMode share_mode = 2;enum DimCheckMode {// STRICT (default) requires that num, channels, height, width each match. STRICT(默认值)要求num、通道、高度和宽度都匹配STRICT = 0;// PERMISSIVE requires only the count (num*channels*height*width) to match. PERMISSIVE只要求count(num*channels*height*width)匹配PERMISSIVE = 1;}// The multiplier on the global learning rate for this parameter.全局学习率的乘数。optional float lr_mult = 3 [default = 1.0];// The multiplier on the global weight decay for this parameter.全局权值的乘数衰减。optional float decay_mult = 4 [default = 1.0];
}// NOTE
// Update the next available ID when you add a new LayerParameter field. 在添加新LayerParameter字段时更新下一个可用ID。
//
// LayerParameter next available layer-specific ID: 149 (last added: clip_param)
message LayerParameter {optional string name = 1; // the layer nameoptional string type = 2; // the layer typerepeated string bottom = 3; // the name of each bottom blobrepeated string top = 4; // the name of each top blob// The train / test phase for computation.optional Phase phase = 10;// The amount of weight to assign each top blob in the objective.// Each layer assigns a default value, usually of either 0 or 1, to each top blob.// 在目标中分配每个顶部blob的权重。每个层为每个顶部blob分配一个默认值，通常为0或1。repeated float loss_weight = 5;// Specifies training parameters (multipliers on global learning constants, and the name and other settings used for weight sharing).// 指定训练参数(全局学习常数的乘数，以及用于权重共享的名称和其他设置)。repeated ParamSpec param = 6;// The blobs containing the numeric parameters of the layer. 包含该层的数值参数的块。repeated BlobProto blobs = 7;// Specifies whether to backpropagate to each bottom. If unspecified, Caffe will automatically infer whether each input needs backpropagation to compute parameter gradients. // If set to true for some inputs, backpropagation to those inputs is forced; if set false for some inputs, backpropagation to those inputs is skipped.// The size must be either 0 or equal to the number of bottoms.// 指定是否反向传播到每个底部。如果未指定，Caffe将自动推断每个输入是否需要反向传播来计算参数梯度。// 如果对某些输入设置为true，则强制反向传播到这些输入;如果某些输入设置为false，则跳过对这些输入的反向传播。大小必须是0或等于底部的数量。repeated bool propagate_down = 11;// Rules controlling whether and when a layer is included in the network,based on the current NetState.  // You may specify a non-zero number of rules to include OR exclude, but not both.  If no include or exclude rules are specified, the layer is always included.  // If the current NetState meets ANY (i.e., one or more) of the specified rules, the layer is included/excluded.// 根据当前网络状态控制网络中是否包含层以及何时包含层的规则。// 你可以指定一个非零数的规则来包含或排除，但不能同时包含或排除。如果没有指定包含或排除规则，则始终包含该层。// 如果当前网络状态满足任何，一个或多个)指定的规则，该层被包括/排除。repeated NetStateRule include = 8;repeated NetStateRule exclude = 9;// Parameters for data pre-processing. 用于数据预处理的参数。optional TransformationParameter transform_param = 100;// Parameters shared by loss layers. 损耗层共享的参数。optional LossParameter loss_param = 101;// Layer type-specific parameters. Layer特定类型参数。//// Note: certain layers may have more than one computational engine for their implementation. // These layers include an Engine type and engine parameter for selecting the implementation.// The default for the engine is set by the ENGINE switch at compile-time.// 注意:某些层的实现可能有多个计算引擎。 这些层包括用于选择实现的引擎类型和引擎参数。 引擎的默认值由引擎开关在编译时设置。optional AccuracyParameter accuracy_param = 102;optional ArgMaxParameter argmax_param = 103;optional BatchNormParameter batch_norm_param = 139;optional BiasParameter bias_param = 141;optional ClipParameter clip_param = 148;optional ConcatParameter concat_param = 104;optional ContrastiveLossParameter contrastive_loss_param = 105;optional ConvolutionParameter convolution_param = 106;optional CropParameter crop_param = 144;optional DataParameter data_param = 107;optional DropoutParameter dropout_param = 108;optional DummyDataParameter dummy_data_param = 109;optional EltwiseParameter eltwise_param = 110;optional ELUParameter elu_param = 140;optional EmbedParameter embed_param = 137;optional ExpParameter exp_param = 111;optional FlattenParameter flatten_param = 135;optional HDF5DataParameter hdf5_data_param = 112;optional HDF5OutputParameter hdf5_output_param = 113;optional HingeLossParameter hinge_loss_param = 114;optional ImageDataParameter image_data_param = 115;optional InfogainLossParameter infogain_loss_param = 116;optional InnerProductParameter inner_product_param = 117;optional InputParameter input_param = 143;optional LogParameter log_param = 134;optional LRNParameter lrn_param = 118;optional MemoryDataParameter memory_data_param = 119;optional MVNParameter mvn_param = 120;optional ParameterParameter parameter_param = 145;optional PoolingParameter pooling_param = 121;optional PowerParameter power_param = 122;optional PReLUParameter prelu_param = 131;optional PythonParameter python_param = 130;optional RecurrentParameter recurrent_param = 146;optional ReductionParameter reduction_param = 136;optional ReLUParameter relu_param = 123;optional ReshapeParameter reshape_param = 133;optional ScaleParameter scale_param = 142;optional SigmoidParameter sigmoid_param = 124;optional SoftmaxParameter softmax_param = 125;optional SPPParameter spp_param = 132;optional SliceParameter slice_param = 126;optional SwishParameter swish_param = 147;optional TanHParameter tanh_param = 127;optional ThresholdParameter threshold_param = 128;optional TileParameter tile_param = 138;optional WindowDataParameter window_data_param = 129;
}