junyuseu / resnet-on-cifar10 Goto Github PK

View Code? Open in Web Editor NEW

129.0 16.0 82.0 7.58 MB

Reimplementation ResNet on cifar10 with caffe

Python 100.00%

resnet cifar10 caffe

resnet-on-cifar10's People

Stargazers

Watchers

Forkers

coderx7 jianweilin deepcompute wait1988 cjsure kevinhuang06 enzo-yang tes3906 rubyway 2php brickjava algpower uerth tangzixia wangsheng1991 devy001 gaffey lubinboooos allan9977 taichu012 lang25150 xinyuegtxy wnylol issac8huxley robustfengbin mingsun-tse zt1112 xugithub1 inkimage flygyyy smartfour yyyyyyx suancaizxt joefannie blackcow dayoncoon andyliu93 imzhou mingchaoxu annie98 signalimagecv xuekuanwang zfxu paul-lipu jinming-su hlqzc2008 jaosonma runauto wzp364 ericeiffel yedideep nixeyjay zonkidd-sh huangchaohuangchao miyaxu0312 gbusr piaomiaoju oneteenager arthur-shi justrypython happyelva lawrencewxj westnight vivienfu i3promise lbbtc shiruipeng1985 pursueorigin ladyrick fantasyjxf xinshaoamoswang wangbaorui bambooyee brother-lee haoji007 666dzy666 miaochenguo frankwanwork ppplang zhaokangjin1996 1343464520 guojy-eatstg

resnet-on-cifar10's Issues

使用自己的数据，训练没问题，测试报错

我把类别改成2，
layer {
name: "InnerProduct1"
type: "InnerProduct"
bottom: "Pooling1"
top: "InnerProduct1"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 1
}
inner_product_param {
num_output: 2
weight_filler {
type: "msra"
}
bias_filler {
type: "constant"
value: 0
}
}
}
在训练的时候不出错，但是测试的时候就报错了这是为啥？

F0810 11:29:23.519547 1856 net.cpp:829] Cannot copy param 0 weights from layer 'InnerProduct1'; shape mismatch. Source param shape is 10 64 (640); target param shape is 2 64 (128). To learn this layer's parameters from scratch rather than copying from a saved net, rename the layer.

what is loss value

I wanna to know if data2 is label?

the meanfile?

mean_file: "../../mean.binaryproto"
crop_size: 28

hi
i have a problem that if meanfile and crop_size used at the same time , is the mean_file usefully?
the train sample size 32x32 mean_file size will be 32x32,how to subtract 28x28

给个建议，不是issuse

你可以吧residual net和palin net的结果画一张图上吗？？
这样分开我也很难作比较啊

Is the pad operation before or after compute mean operation?

Hi,
I want to know is the pad operation before or after computing mean operation?
Thank you~

预测网络没输出.没输出

name: "resnet_cifar10"
layer {
  name: "data"
  type: "Input"
  top: "Data1"
  top: "Data2"
  input_param { shape: { dim: 1 dim: 3 dim: 28 dim: 28 } }
}


layer {
  name: "Convolution1"
  type: "Convolution"
  bottom: "Data1"
  top: "Convolution1"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm1"
  type: "BatchNorm"
  bottom: "Convolution1"
  top: "Convolution1"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale1"
  type: "Scale"
  bottom: "Convolution1"
  top: "Convolution1"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU1"
  type: "ReLU"
  bottom: "Convolution1"
  top: "Convolution1"
}
layer {
  name: "Convolution2"
  type: "Convolution"
  bottom: "Convolution1"
  top: "Convolution2"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm2"
  type: "BatchNorm"
  bottom: "Convolution2"
  top: "Convolution2"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale2"
  type: "Scale"
  bottom: "Convolution2"
  top: "Convolution2"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU2"
  type: "ReLU"
  bottom: "Convolution2"
  top: "Convolution2"
}
layer {
  name: "Convolution3"
  type: "Convolution"
  bottom: "Convolution2"
  top: "Convolution3"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm3"
  type: "BatchNorm"
  bottom: "Convolution3"
  top: "Convolution3"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale3"
  type: "Scale"
  bottom: "Convolution3"
  top: "Convolution3"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise1"
  type: "Eltwise"
  bottom: "Convolution1"
  bottom: "Convolution3"
  top: "Eltwise1"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU3"
  type: "ReLU"
  bottom: "Eltwise1"
  top: "Eltwise1"
}
layer {
  name: "Convolution4"
  type: "Convolution"
  bottom: "Eltwise1"
  top: "Convolution4"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm4"
  type: "BatchNorm"
  bottom: "Convolution4"
  top: "Convolution4"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale4"
  type: "Scale"
  bottom: "Convolution4"
  top: "Convolution4"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU4"
  type: "ReLU"
  bottom: "Convolution4"
  top: "Convolution4"
}
layer {
  name: "Convolution5"
  type: "Convolution"
  bottom: "Convolution4"
  top: "Convolution5"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm5"
  type: "BatchNorm"
  bottom: "Convolution5"
  top: "Convolution5"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale5"
  type: "Scale"
  bottom: "Convolution5"
  top: "Convolution5"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise2"
  type: "Eltwise"
  bottom: "Eltwise1"
  bottom: "Convolution5"
  top: "Eltwise2"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU5"
  type: "ReLU"
  bottom: "Eltwise2"
  top: "Eltwise2"
}
layer {
  name: "Convolution6"
  type: "Convolution"
  bottom: "Eltwise2"
  top: "Convolution6"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm6"
  type: "BatchNorm"
  bottom: "Convolution6"
  top: "Convolution6"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale6"
  type: "Scale"
  bottom: "Convolution6"
  top: "Convolution6"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU6"
  type: "ReLU"
  bottom: "Convolution6"
  top: "Convolution6"
}
layer {
  name: "Convolution7"
  type: "Convolution"
  bottom: "Convolution6"
  top: "Convolution7"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 16
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.118
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm7"
  type: "BatchNorm"
  bottom: "Convolution7"
  top: "Convolution7"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale7"
  type: "Scale"
  bottom: "Convolution7"
  top: "Convolution7"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise3"
  type: "Eltwise"
  bottom: "Eltwise2"
  bottom: "Convolution7"
  top: "Eltwise3"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU7"
  type: "ReLU"
  bottom: "Eltwise3"
  top: "Eltwise3"
}
layer {
  name: "Convolution8"
  type: "Convolution"
  bottom: "Eltwise3"
  top: "Convolution8"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 0
    kernel_size: 1
    stride: 2
    weight_filler {
      type: "gaussian"
      std: 0.25
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm8"
  type: "BatchNorm"
  bottom: "Convolution8"
  top: "Convolution8"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale8"
  type: "Scale"
  bottom: "Convolution8"
  top: "Convolution8"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Convolution9"
  type: "Convolution"
  bottom: "Eltwise3"
  top: "Convolution9"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 2
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm9"
  type: "BatchNorm"
  bottom: "Convolution9"
  top: "Convolution9"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale9"
  type: "Scale"
  bottom: "Convolution9"
  top: "Convolution9"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU8"
  type: "ReLU"
  bottom: "Convolution9"
  top: "Convolution9"
}
layer {
  name: "Convolution10"
  type: "Convolution"
  bottom: "Convolution9"
  top: "Convolution10"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm10"
  type: "BatchNorm"
  bottom: "Convolution10"
  top: "Convolution10"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale10"
  type: "Scale"
  bottom: "Convolution10"
  top: "Convolution10"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise4"
  type: "Eltwise"
  bottom: "Convolution8"
  bottom: "Convolution10"
  top: "Eltwise4"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU9"
  type: "ReLU"
  bottom: "Eltwise4"
  top: "Eltwise4"
}
layer {
  name: "Convolution11"
  type: "Convolution"
  bottom: "Eltwise4"
  top: "Convolution11"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm11"
  type: "BatchNorm"
  bottom: "Convolution11"
  top: "Convolution11"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale11"
  type: "Scale"
  bottom: "Convolution11"
  top: "Convolution11"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU10"
  type: "ReLU"
  bottom: "Convolution11"
  top: "Convolution11"
}
layer {
  name: "Convolution12"
  type: "Convolution"
  bottom: "Convolution11"
  top: "Convolution12"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm12"
  type: "BatchNorm"
  bottom: "Convolution12"
  top: "Convolution12"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale12"
  type: "Scale"
  bottom: "Convolution12"
  top: "Convolution12"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise5"
  type: "Eltwise"
  bottom: "Eltwise4"
  bottom: "Convolution12"
  top: "Eltwise5"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU11"
  type: "ReLU"
  bottom: "Eltwise5"
  top: "Eltwise5"
}
layer {
  name: "Convolution13"
  type: "Convolution"
  bottom: "Eltwise5"
  top: "Convolution13"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm13"
  type: "BatchNorm"
  bottom: "Convolution13"
  top: "Convolution13"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale13"
  type: "Scale"
  bottom: "Convolution13"
  top: "Convolution13"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU12"
  type: "ReLU"
  bottom: "Convolution13"
  top: "Convolution13"
}
layer {
  name: "Convolution14"
  type: "Convolution"
  bottom: "Convolution13"
  top: "Convolution14"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 32
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.083
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm14"
  type: "BatchNorm"
  bottom: "Convolution14"
  top: "Convolution14"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale14"
  type: "Scale"
  bottom: "Convolution14"
  top: "Convolution14"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise6"
  type: "Eltwise"
  bottom: "Eltwise5"
  bottom: "Convolution14"
  top: "Eltwise6"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU13"
  type: "ReLU"
  bottom: "Eltwise6"
  top: "Eltwise6"
}
layer {
  name: "Convolution15"
  type: "Convolution"
  bottom: "Eltwise6"
  top: "Convolution15"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 0
    kernel_size: 1
    stride: 2
    weight_filler {
      type: "gaussian"
      std: 0.176776695297
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm15"
  type: "BatchNorm"
  bottom: "Convolution15"
  top: "Convolution15"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale15"
  type: "Scale"
  bottom: "Convolution15"
  top: "Convolution15"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Convolution16"
  type: "Convolution"
  bottom: "Eltwise6"
  top: "Convolution16"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 2
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm16"
  type: "BatchNorm"
  bottom: "Convolution16"
  top: "Convolution16"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale16"
  type: "Scale"
  bottom: "Convolution16"
  top: "Convolution16"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU14"
  type: "ReLU"
  bottom: "Convolution16"
  top: "Convolution16"
}
layer {
  name: "Convolution17"
  type: "Convolution"
  bottom: "Convolution16"
  top: "Convolution17"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm17"
  type: "BatchNorm"
  bottom: "Convolution17"
  top: "Convolution17"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale17"
  type: "Scale"
  bottom: "Convolution17"
  top: "Convolution17"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise7"
  type: "Eltwise"
  bottom: "Convolution15"
  bottom: "Convolution17"
  top: "Eltwise7"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU15"
  type: "ReLU"
  bottom: "Eltwise7"
  top: "Eltwise7"
}
layer {
  name: "Convolution18"
  type: "Convolution"
  bottom: "Eltwise7"
  top: "Convolution18"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm18"
  type: "BatchNorm"
  bottom: "Convolution18"
  top: "Convolution18"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale18"
  type: "Scale"
  bottom: "Convolution18"
  top: "Convolution18"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU16"
  type: "ReLU"
  bottom: "Convolution18"
  top: "Convolution18"
}
layer {
  name: "Convolution19"
  type: "Convolution"
  bottom: "Convolution18"
  top: "Convolution19"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm19"
  type: "BatchNorm"
  bottom: "Convolution19"
  top: "Convolution19"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale19"
  type: "Scale"
  bottom: "Convolution19"
  top: "Convolution19"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise8"
  type: "Eltwise"
  bottom: "Eltwise7"
  bottom: "Convolution19"
  top: "Eltwise8"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU17"
  type: "ReLU"
  bottom: "Eltwise8"
  top: "Eltwise8"
}
layer {
  name: "Convolution20"
  type: "Convolution"
  bottom: "Eltwise8"
  top: "Convolution20"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm20"
  type: "BatchNorm"
  bottom: "Convolution20"
  top: "Convolution20"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale20"
  type: "Scale"
  bottom: "Convolution20"
  top: "Convolution20"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "ReLU18"
  type: "ReLU"
  bottom: "Convolution20"
  top: "Convolution20"
}
layer {
  name: "Convolution21"
  type: "Convolution"
  bottom: "Convolution20"
  top: "Convolution21"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 0
  }
  convolution_param {
    num_output: 64
    pad: 1
    kernel_size: 3
    stride: 1
    weight_filler {
      type: "gaussian"
      std: 0.059
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
layer {
  name: "BatchNorm21"
  type: "BatchNorm"
  bottom: "Convolution21"
  top: "Convolution21"
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
  param {
    lr_mult: 0
    decay_mult: 0
  }
}
layer {
  name: "Scale21"
  type: "Scale"
  bottom: "Convolution21"
  top: "Convolution21"
  scale_param {
    bias_term: true
  }
}
layer {
  name: "Eltwise9"
  type: "Eltwise"
  bottom: "Eltwise8"
  bottom: "Convolution21"
  top: "Eltwise9"
  eltwise_param {
    operation: SUM
  }
}
layer {
  name: "ReLU19"
  type: "ReLU"
  bottom: "Eltwise9"
  top: "Eltwise9"
}
layer {
  name: "Pooling1"
  type: "Pooling"
  bottom: "Eltwise9"
  top: "Pooling1"
  pooling_param {
    pool: AVE
    global_pooling: true
  }
}
layer {
  name: "InnerProduct1"
  type: "InnerProduct"
  bottom: "Pooling1"
  top: "InnerProduct1"
  param {
    lr_mult: 1
    decay_mult: 1
  }
  param {
    lr_mult: 2
    decay_mult: 1
  }
  inner_product_param {
    num_output: 10
    weight_filler {
      type: "gaussian"
      std: 0.01
    }
    bias_filler {
      type: "constant"
      value: 0
    }
  }
}
#layer {
#  name: "SoftmaxWithLoss1"
#  type: "SoftmaxWithLoss"
#  bottom: "InnerProduct1"
#  bottom: "Data2"
#  top: "SoftmaxWithLoss1"
#}
layer {
    bottom: "InnerProduct1"
    top: "prob"
    name: "prob"
    type: "Softmax"
}

What is model for Resnet 110

Hello, It is very interesting in your result. However, up to 110 layers, they often use identity skip connections in "Identity Mappings in Deep Residual Networks". They also show very performance using the network. Do you try to implement it in caffe? Thanks

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.