eric612 / caffe-yolov3-windows Goto Github PK
View Code? Open in Web Editor NEWA windows caffe implementation of YOLO detection network
License: Other
caffe-yolov3-windows's Introduction
caffe-yolov3-windows's People
Contributors
Stargazers
Watchers
Forkers
feitiandemiaomi woyaofeixiang liqidong aa12356jm jiancunwang satchelwu hklee2040 mypostw xingyewuyu jnulzl lvpchen tonychouzju baiyu7297 xyt2008 codermckee decenwill huipengzhang jiayanyuan baifanysu dengkee yiran-thu lqs19881030 joeysu anhuipl2010 youngbaby123 labimage aaronyking facegen kk52099 lsps2009 wanjinchang styjb cxf2015 linquanxu zhukkang tencyrush laoyingbu w510056105 wuxiaolianggit alixing calvinpaean miwaliu koala-good amirunpri2018 fengzhizi319 jacksonl1 aqsc runauto yxliang zp1018 chop2 duducode aiyangyang963 jrznbu haiyang21 isaiah-li dragonzly leo-xxx lovehell jacke121 njustghw monoloxo tim-lee-cn kingsleyzw zhang11wu4 dltensor jinjunkuang jiaerwang0328 randal7 lubinboooos pjhardy we0091234 ylh071032 betterhalfwzm lionelleee danigy zilipeng dangxusheng icprog liuwenhaha orzorzorzxx rocinant 18669976234 zyc4me zhengjunqiang wavelet2008 xiaozhishang multierection p1b234 hito0512 jocooob wxs29 leeyou glgoo mzpmzk liulangxing asukaj lanxielee hedilong haozhang-qdcaffe-yolov3-windows's Issues
question about platform
Can I switch to Linux platform training and using?
yolov3 lite训练的一些疑惑
1.batch size设为6，是不是太小了？对caffe不熟，batch size=6,iter_size=4,那么每次训练的样本数是int(6/4)?
2.多个resize_param，到底用的哪一个？是不是没有实现多尺度训练？我看显存占用一直没变。
nvcc fatal : Unsupported gpu architecture 'compute_90'
Building NVCC (Device) object src/caffe/CMakeFiles/cuda_compile_1.dir/layers/Release/cuda_compile_1_generated_absval_layer.cu.obj
nvcc fatal : Unsupported gpu architecture 'compute_90'
what should i do?
Why X&Y&HasObject&classlabel loss not use binary_crossentropy?
your W&H loss use MSE is as same as paper,but your X&Y&HasObject&classlabel also use MSE is not as same as paper,paper use binary_crossentropy，why?
yolo_v2_loss layer
Can you show me a prototxt example about how to set yolo_v2_loss layer parameters?
one question
[libprotobuf ERROR C:\Users\guillaume\work\caffe-builder\build_v140_x64\packages\protobuf\protobuf_download-prefix\src\protobuf_download\src\google\protobuf\text_format.cc:298] Error parsing text-format caffe.NetParameter: 27:18: Non-repeated field "resize_param" is specified multiple times.
F0915 11:44:01.675608 2252 upgrade_proto.cpp:90] Check failed: ReadProtoFromTextFile(param_file, param) Failed to parse NetParameter file: mobilenet_yolov3_train.prototxt
这是为什么呢？
Download link for LMDB is no longer valid.
Hello,
I am trying to follow the readme instruction. However, the download link for LMDB does not work any more. Could you please fix the link?
Many thanks,
Eungi Hong
Question about imagedemoOpenCV
Hi I really appreciate what you did.
I successfully built this project, but when I ran "demo_yolo_lite.cmd", I got the following error:
- "imagedemoOpenCV Error: Requested object was not found (could not open directory: $RECYCLE.BIN\S-1-5-18) in glob_rec, file C:\Users\guillaume\work\caffe-builder\build_v140_x64\packages\opencv\opencv_download-prefix\src\opencv_download\modules\core\src\glob.cpp, line 257"
Could you help me? Thanks a lot.
[Training issue]为什么在某些分辨率下，算法会对一类目标漏检？
up主你好，感谢你的工作。
1.我最近在用一些自己的数据训练Yolov3_lite，然后进行测试。因为数据图像的尺度（1280*1280）固定，目标尺寸大小固定，所以我重新更改了网络结构，并用
It is very strange , can you try this way ?
的办法重新计算了anchor boxes，但是检测结果不理想，在测试集中该类目标一个都没检测出来！其他目标检测结果都正常；
2.此外，我还试过将图像分辨率（1024*1024）改小，并重新计算anchor，结果在该分辨率下，目标能检测出来，但是置信度较低。
3.若是将图像分辨率改为832*832，则该类目标检测结果正常。
My train prototxt
name: "MobileNet-YOLO"
layer {
name: "data"
type: "AnnotatedData"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
scale: 0.007843
mirror: true
mean_value: 127.5
mean_value: 127.5
mean_value: 127.5
resize_param {
prob: 1.0
resize_mode: WARP
height: 1280
width: 1280
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
emit_constraint {
emit_type: CENTER
}
distort_param {
brightness_prob: 0.5
brightness_delta: 32.0
contrast_prob: 0.5
contrast_lower: 0.5
contrast_upper: 1.5
hue_prob: 0.5
hue_delta: 18.0
saturation_prob: 0.5
saturation_lower: 0.5
saturation_upper: 1.5
random_order_prob: 0.0
}
expand_param {
prob: 0.5
max_expand_ratio: 3.0
}
}
data_param {
source: "examples/VOC0712/VOC0712_trainval_lmdb"
batch_size: 1
backend: LMDB
}
annotated_data_param {
yolo_data_type : 1
yolo_data_jitter : 0.3
label_map_file: "data/VOC0712/labelmap_voc.prototxt"
}
}
layer {
name: "conv0"
type: "Convolution"
bottom: "data"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
stride: 2
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv0/bn"
type: "BatchNorm"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv0/scale"
type: "Scale"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv0/relu"
type: "ReLU"
bottom: "conv0"
top: "conv0"
}
layer {
name: "conv1/dw"
type: "DepthwiseConvolution"
bottom: "conv0"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
group: 32
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/dw/bn"
type: "BatchNorm"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/dw/scale"
type: "Scale"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/dw/relu"
type: "ReLU"
bottom: "conv1/dw"
top: "conv1/dw"
}
layer {
name: "conv1"
type: "Convolution"
bottom: "conv1/dw"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/bn"
type: "BatchNorm"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/scale"
type: "Scale"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/relu"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv2/dw"
type: "DepthwiseConvolution"
bottom: "conv1"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 64
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/dw/bn"
type: "BatchNorm"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/dw/scale"
type: "Scale"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/dw/relu"
type: "ReLU"
bottom: "conv2/dw"
top: "conv2/dw"
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv2/dw"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/bn"
type: "BatchNorm"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/scale"
type: "Scale"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/relu"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv3/dw"
type: "DepthwiseConvolution"
bottom: "conv2"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/dw/bn"
type: "BatchNorm"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/dw/scale"
type: "Scale"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/dw/relu"
type: "ReLU"
bottom: "conv3/dw"
top: "conv3/dw"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv3/dw"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/bn"
type: "BatchNorm"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/scale"
type: "Scale"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/relu"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4/dw"
type: "DepthwiseConvolution"
bottom: "conv3"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/dw/bn"
type: "BatchNorm"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/dw/scale"
type: "Scale"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/dw/relu"
type: "ReLU"
bottom: "conv4/dw"
top: "conv4/dw"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv4/dw"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/bn"
type: "BatchNorm"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/scale"
type: "Scale"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/relu"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5/dw"
type: "DepthwiseConvolution"
bottom: "conv4"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/dw/bn"
type: "BatchNorm"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/dw/scale"
type: "Scale"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/dw/relu"
type: "ReLU"
bottom: "conv5/dw"
top: "conv5/dw"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv5/dw"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/bn"
type: "BatchNorm"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/scale"
type: "Scale"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/relu"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "conv6/dw"
type: "DepthwiseConvolution"
bottom: "conv5"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/dw/bn"
type: "BatchNorm"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/dw/scale"
type: "Scale"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/dw/relu"
type: "ReLU"
bottom: "conv6/dw"
top: "conv6/dw"
}
layer {
name: "conv6"
type: "Convolution"
bottom: "conv6/dw"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/bn"
type: "BatchNorm"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/scale"
type: "Scale"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/relu"
type: "ReLU"
bottom: "conv6"
top: "conv6"
}
layer {
name: "conv7/dw"
type: "DepthwiseConvolution"
bottom: "conv6"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/dw/bn"
type: "BatchNorm"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/dw/scale"
type: "Scale"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/dw/relu"
type: "ReLU"
bottom: "conv7/dw"
top: "conv7/dw"
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv7/dw"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/bn"
type: "BatchNorm"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/scale"
type: "Scale"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/relu"
type: "ReLU"
bottom: "conv7"
top: "conv7"
}
layer {
name: "conv8/dw"
type: "DepthwiseConvolution"
bottom: "conv7"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/dw/bn"
type: "BatchNorm"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/dw/scale"
type: "Scale"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/dw/relu"
type: "ReLU"
bottom: "conv8/dw"
top: "conv8/dw"
}
layer {
name: "conv8"
type: "Convolution"
bottom: "conv8/dw"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/bn"
type: "BatchNorm"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/scale"
type: "Scale"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/relu"
type: "ReLU"
bottom: "conv8"
top: "conv8"
}
layer {
name: "conv9/dw"
type: "DepthwiseConvolution"
bottom: "conv8"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/dw/bn"
type: "BatchNorm"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/dw/scale"
type: "Scale"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/dw/relu"
type: "ReLU"
bottom: "conv9/dw"
top: "conv9/dw"
}
layer {
name: "conv9"
type: "Convolution"
bottom: "conv9/dw"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/bn"
type: "BatchNorm"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/scale"
type: "Scale"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/relu"
type: "ReLU"
bottom: "conv9"
top: "conv9"
}
layer {
name: "conv10/dw"
type: "DepthwiseConvolution"
bottom: "conv9"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/dw/bn"
type: "BatchNorm"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/dw/scale"
type: "Scale"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/dw/relu"
type: "ReLU"
bottom: "conv10/dw"
top: "conv10/dw"
}
layer {
name: "conv10"
type: "Convolution"
bottom: "conv10/dw"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/bn"
type: "BatchNorm"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/scale"
type: "Scale"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/relu"
type: "ReLU"
bottom: "conv10"
top: "conv10"
}
layer {
name: "conv11/dw"
type: "DepthwiseConvolution"
bottom: "conv10"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/dw/bn"
type: "BatchNorm"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/dw/scale"
type: "Scale"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/dw/relu"
type: "ReLU"
bottom: "conv11/dw"
top: "conv11/dw"
}
layer {
name: "conv11"
type: "Convolution"
bottom: "conv11/dw"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/bn"
type: "BatchNorm"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/scale"
type: "Scale"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/relu"
type: "ReLU"
bottom: "conv11"
top: "conv11"
}
layer {
name: "conv12/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/dw/bn"
type: "BatchNorm"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/dw/scale"
type: "Scale"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/dw/relu"
type: "ReLU"
bottom: "conv12/dw"
top: "conv12/dw"
}
layer {
name: "conv12"
type: "Convolution"
bottom: "conv12/dw"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/bn"
type: "BatchNorm"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/scale"
type: "Scale"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/relu"
type: "ReLU"
bottom: "conv12"
top: "conv12"
}
layer {
name: "conv13/dw"
type: "DepthwiseConvolution"
bottom: "conv12"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/dw/bn"
type: "BatchNorm"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/dw/scale"
type: "Scale"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/dw/relu"
type: "ReLU"
bottom: "conv13/dw"
top: "conv13/dw"
}
layer {
name: "conv13"
type: "Convolution"
bottom: "conv13/dw"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/bn"
type: "BatchNorm"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/scale"
type: "Scale"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/relu"
type: "ReLU"
bottom: "conv13"
top: "conv13"
}
layer {
name: "conv16/dw"
type: "DepthwiseConvolution"
bottom: "conv13"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/dw/bn"
type: "BatchNorm"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/dw/scale"
type: "Scale"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/dw/relu"
type: "ReLU"
bottom: "conv16/dw"
top: "conv16/dw"
}
layer {
name: "conv16"
type: "Convolution"
bottom: "conv16/dw"
top: "conv16"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/bn"
type: "BatchNorm"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/scale"
type: "Scale"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/relu"
type: "ReLU"
bottom: "conv16"
top: "conv16"
}
layer {
name: "upsample"
type: "Deconvolution"
bottom: "conv16"
top: "upsample"
param { lr_mult: 0 decay_mult: 0 }
convolution_param {
num_output: 512
kernel_size: 4 stride: 2 pad: 1
group: 512
weight_filler: { type: "bilinear" }
bias_term: false
}
}
layer {
name: "conv17/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/dw/bn"
type: "BatchNorm"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/dw/scale"
type: "Scale"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/dw/relu"
type: "ReLU"
bottom: "conv17/dw"
top: "conv17/dw"
}
layer {
name: "conv17"
type: "Convolution"
bottom: "conv17/dw"
top: "conv17"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/bn"
type: "BatchNorm"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/scale"
type: "Scale"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/relu"
type: "ReLU"
bottom: "conv17"
top: "conv17"
}
layer {
name: "conv17/sum"
type: "Eltwise"
bottom: "conv17"
bottom: "upsample"
top: "conv17/sum"
}
layer {
name: "conv18/dw"
type: "DepthwiseConvolution"
bottom: "conv17/sum"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/dw/bn"
type: "BatchNorm"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/dw/scale"
type: "Scale"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/dw/relu"
type: "ReLU"
bottom: "conv18/dw"
top: "conv18/dw"
}
layer {
name: "conv18"
type: "Convolution"
bottom: "conv18/dw"
top: "conv18"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/bn"
type: "BatchNorm"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/scale"
type: "Scale"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/relu"
type: "ReLU"
bottom: "conv18"
top: "conv18"
}
layer {
name: "conv20_new"
type: "Convolution"
bottom: "conv16"
top: "conv20_new"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 1024
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "conv20_new/bn"
type: "BatchNorm"
bottom: "conv20_new"
top: "conv20_new"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv20_new/scale"
type: "Scale"
bottom: "conv20_new"
top: "conv20_new"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv20_new/relu"
type: "ReLU"
bottom: "conv20_new"
top: "conv20_new"
}
layer {
name: "conv21"
type: "Convolution"
bottom: "conv18"
top: "conv21"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "conv22/dw"
type: "DepthwiseConvolution"
bottom: "conv20_new"
top: "conv22/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: true
pad: 1
kernel_size: 3
stride: 2
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv22/dw/bn"
type: "BatchNorm"
bottom: "conv22/dw"
top: "conv22/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv22/dw/scale"
type: "Scale"
bottom: "conv22/dw"
top: "conv22/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv22/dw/relu"
type: "ReLU"
bottom: "conv22/dw"
top: "conv22/dw"
}
layer {
name: "conv22"
type: "Convolution"
bottom: "conv22/dw"
top: "conv22"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: true
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv22/bn"
type: "BatchNorm"
bottom: "conv22"
top: "conv22"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv22/scale"
type: "Scale"
bottom: "conv22"
top: "conv22"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv22/relu"
type: "ReLU"
bottom: "conv22"
top: "conv22"
}
layer {
name: "conv23/dw"
type: "DepthwiseConvolution"
bottom: "conv22"
top: "conv23/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: true
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv23/dw/bn"
type: "BatchNorm"
bottom: "conv23/dw"
top: "conv23/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv23/dw/scale"
type: "Scale"
bottom: "conv23/dw"
top: "conv23/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv23/dw/relu"
type: "ReLU"
bottom: "conv23/dw"
top: "conv23/dw"
}
layer {
name: "conv23"
type: "Convolution"
bottom: "conv23/dw"
top: "conv23"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 75
bias_term: true
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv20_old"
type: "Convolution"
bottom: "conv20_new"
top: "conv20_old"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "Yolov3Loss1"
type: "Yolov3"
bottom: "conv23"
bottom: "label"
top: "det_loss1"
loss_weight: 1
yolov3_param {
side: 13
num_class: 20
num: 3
object_scale: 5.0
noobject_scale: 1.0
class_scale: 1.0
coord_scale: 1.0
thresh: 0.6
anchors_scale : 64
use_logic_gradient : false
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 34
biases: 43
biases: 66
biases: 51
biases: 54
biases: 76
biases: 183
biases: 84
biases: 128
biases: 142
biases: 317
biases: 270
biases: 77
biases: 85
biases: 85
biases: 104
biases: 100
biases: 116
mask:3
mask:4
mask:5
}
}
layer {
name: "Yolov3Loss2"
type: "Yolov3"
bottom: "conv21"
bottom: "label"
top: "det_loss2"
loss_weight: 1
yolov3_param {
side: 26
num_class: 20
num: 3
object_scale: 5.0
noobject_scale: 1.0
class_scale: 1.0
coord_scale: 1.0
thresh: 0.6
anchors_scale : 16
use_logic_gradient : false
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 34
biases: 43
biases: 66
biases: 51
biases: 54
biases: 76
biases: 183
biases: 84
biases: 128
biases: 142
biases: 317
biases: 270
biases: 77
biases: 85
biases: 85
biases: 104
biases: 100
biases: 116
mask:0
mask:1
mask:2
}
}
layer {
name: "Yolov3Loss3"
type: "Yolov3"
bottom: "conv20_old"
bottom: "label"
top: "det_loss3"
loss_weight: 1
yolov3_param {
side: 13
num_class: 20
num: 3
object_scale: 5.0
noobject_scale: 1.0
class_scale: 1.0
coord_scale: 1.0
thresh: 0.6
anchors_scale : 32
use_logic_gradient : false
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 34
biases: 43
biases: 66
biases: 51
biases: 54
biases: 76
biases: 183
biases: 84
biases: 128
biases: 142
biases: 317
biases: 270
biases: 77
biases: 85
biases: 85
biases: 104
biases: 100
biases: 116
mask:6
mask:7
mask:8
}
}
请问下，为什么会出现这种情况？
如果我想要在1280*1280分辨率下正确检测出该目标，应该怎么做？
build_win.cmd编译失败
CMake Error at cuda_compile_1_generated_absval_layer.cu.obj.Release.cmake:220 (message):
Error generating
H:/Project/caffe-source/Caffe-YOLOv3-Windows-master/build/src/caffe/CMakeFiles/cuda_compile_1.dir/layers/Release/cuda_compile_1_generated_absval_layer.cu.obj
ERROR: Build failed
@eric612 您好，我编译的时候出现上面的错误，尝试了半天也没有找到很好的解决办法，想请教一下您，这是什么原因导致的呢？
How to generate train data
@eric612 Hi，you only provide lmdb data. But I want to know how to prepare my own train data？
typedef in bbox_util.hpp
where PriorBoxParameter_CodeType， EmitConstraint_EmitType， MultiBoxLossParameter_MatchType and some other classes are declared in bbox_util.hpp. thank you!
build caffe-yolov3-windows with visual studio 2015 fail!
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\convert_imageset.vcxproj”(默认目标) (15) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\convert_imageset.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\convert_mnist_data.vcxproj”(默认目标) (17) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\convert_mnist_data.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\convert_mnist_siamese_data.vcxproj”(默认目标) (19) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\convert_mnist_siamese_data.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\create_label_map.vcxproj”(默认目标) (21) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\create_label_map.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\device_query.vcxproj”(默认目标) (23) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\device_query.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\extract_features.vcxproj”(默认目标) (25) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\extract_features.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\finetune_net.vcxproj”(默认目标) (27) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\finetune_net.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\get_image_size.vcxproj”(默认目标) (29) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\get_image_size.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\net_speed_benchmark.vcxproj”(默认目标) (31) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\net_speed_benchmark.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\python\pycaffe.vcxproj”(默认目标) (33) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\python\pycaffe.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\ssd_detect.vcxproj”(默认目标) (35) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\examples\ssd_detect.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\test_net.vcxproj”(默认目标) (37) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\test_net.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\train_net.vcxproj”(默认目标) (39) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\train_net.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_net_proto_binary.vcxproj”(默认目标) (41) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_net_proto_binary.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_net_proto_text.vcxproj”(默认目标) (43) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_net_proto_text.vcxproj]
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\ALL_BUILD.vcxproj”(默认目标) (1) ->
“E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_solver_proto_text.vcxproj”(默认目标) (45) ->
LINK : fatal error LNK1181: 无法打开输入文件“\openmp.obj” [E:\Workspace\Project\trunk\Caffe-YOLOv3-Windows-master\build\tools\upgrade_solver_proto_text.vcxproj]
10439 个警告
21 个错误
已用时间 00:18:39.43
ERROR: Build failed
training is ok ,deploy test is ok ,but not get good result .
hello ! i am use mobilenet_yolov3_lite_solver.prototxt ,mobilenet_yolov3_lite_train.prototxt and mobilenet_yolov3_lite_test.prototxt to get the caffe_model_file.
train_net: "models/yolov3/mobilenet_yolov3_lite_train.prototxt"
test_net: "models/yolov3/mobilenet_yolov3_lite_test.prototxt"
test_iter: 4952
test_interval: 1000
base_lr: 0.0005
display: 10
max_iter: 120000
lr_policy: "multistep"
gamma: 0.5
weight_decay: 0.00005
snapshot: 1000
snapshot_prefix: "models/yolov3/mobilenet_yolov3_lite_deploy"
solver_mode: GPU
debug_info: false
snapshot_after_train: true
test_initialization: false
average_loss: 10
stepvalue: 20000
stepvalue: 40000
stepvalue: 60000
iter_size: 4
type: "RMSProp"
eval_type: "detection"
ap_version: "11point"
show_per_class_result: true
name: "MobileNet-YOLO"
layer {
name: "data"
type: "AnnotatedData"
top: "data"
top: "label"
include {
phase: TRAIN
}
transform_param {
scale: 0.007843
mirror: true
mean_value: 127.5
mean_value: 127.5
mean_value: 127.5
resize_param {
prob: 0.1
resize_mode: WARP
height: 608
width: 608
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 416
width: 416
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 320
width: 320
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 352
width: 352
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 384
width: 384
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 448
width: 448
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 480
width: 480
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 512
width: 512
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 544
width: 544
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
resize_param {
prob: 0.1
resize_mode: WARP
height: 576
width: 576
interp_mode: LINEAR
interp_mode: AREA
interp_mode: LANCZOS4
}
emit_constraint {
emit_type: CENTER
}
distort_param {
brightness_prob: 0.5
brightness_delta: 32.0
contrast_prob: 0.5
contrast_lower: 0.5
contrast_upper: 1.5
hue_prob: 0.5
hue_delta: 18.0
saturation_prob: 0.5
saturation_lower: 0.5
saturation_upper: 1.5
random_order_prob: 0.0
}
expand_param {
prob: 0.5
max_expand_ratio: 3.0
}
}
data_param {
source: "examples/VOC0712/VOC0712_trainval_lmdb"
batch_size: 6
backend: LMDB
}
annotated_data_param {
yolo_data_type : 1
yolo_data_jitter : 0.3
label_map_file: "data/VOC0712/labelmap_voc.prototxt"
}
}
layer {
name: "conv0"
type: "Convolution"
bottom: "data"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
stride: 2
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv0/bn"
type: "BatchNorm"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv0/scale"
type: "Scale"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv0/relu"
type: "ReLU"
bottom: "conv0"
top: "conv0"
}
layer {
name: "conv1/dw"
type: "DepthwiseConvolution"
bottom: "conv0"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
group: 32
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/dw/bn"
type: "BatchNorm"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/dw/scale"
type: "Scale"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/dw/relu"
type: "ReLU"
bottom: "conv1/dw"
top: "conv1/dw"
}
layer {
name: "conv1"
type: "Convolution"
bottom: "conv1/dw"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/bn"
type: "BatchNorm"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/scale"
type: "Scale"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/relu"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv2/dw"
type: "DepthwiseConvolution"
bottom: "conv1"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 64
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/dw/bn"
type: "BatchNorm"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/dw/scale"
type: "Scale"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/dw/relu"
type: "ReLU"
bottom: "conv2/dw"
top: "conv2/dw"
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv2/dw"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/bn"
type: "BatchNorm"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/scale"
type: "Scale"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/relu"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv3/dw"
type: "DepthwiseConvolution"
bottom: "conv2"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/dw/bn"
type: "BatchNorm"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/dw/scale"
type: "Scale"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/dw/relu"
type: "ReLU"
bottom: "conv3/dw"
top: "conv3/dw"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv3/dw"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/bn"
type: "BatchNorm"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/scale"
type: "Scale"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/relu"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4/dw"
type: "DepthwiseConvolution"
bottom: "conv3"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/dw/bn"
type: "BatchNorm"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/dw/scale"
type: "Scale"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/dw/relu"
type: "ReLU"
bottom: "conv4/dw"
top: "conv4/dw"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv4/dw"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/bn"
type: "BatchNorm"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/scale"
type: "Scale"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/relu"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5/dw"
type: "DepthwiseConvolution"
bottom: "conv4"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/dw/bn"
type: "BatchNorm"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/dw/scale"
type: "Scale"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/dw/relu"
type: "ReLU"
bottom: "conv5/dw"
top: "conv5/dw"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv5/dw"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/bn"
type: "BatchNorm"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/scale"
type: "Scale"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/relu"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "conv6/dw"
type: "DepthwiseConvolution"
bottom: "conv5"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/dw/bn"
type: "BatchNorm"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/dw/scale"
type: "Scale"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/dw/relu"
type: "ReLU"
bottom: "conv6/dw"
top: "conv6/dw"
}
layer {
name: "conv6"
type: "Convolution"
bottom: "conv6/dw"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/bn"
type: "BatchNorm"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/scale"
type: "Scale"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/relu"
type: "ReLU"
bottom: "conv6"
top: "conv6"
}
layer {
name: "conv7/dw"
type: "DepthwiseConvolution"
bottom: "conv6"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/dw/bn"
type: "BatchNorm"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/dw/scale"
type: "Scale"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/dw/relu"
type: "ReLU"
bottom: "conv7/dw"
top: "conv7/dw"
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv7/dw"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/bn"
type: "BatchNorm"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/scale"
type: "Scale"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/relu"
type: "ReLU"
bottom: "conv7"
top: "conv7"
}
layer {
name: "conv8/dw"
type: "DepthwiseConvolution"
bottom: "conv7"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/dw/bn"
type: "BatchNorm"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/dw/scale"
type: "Scale"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/dw/relu"
type: "ReLU"
bottom: "conv8/dw"
top: "conv8/dw"
}
layer {
name: "conv8"
type: "Convolution"
bottom: "conv8/dw"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/bn"
type: "BatchNorm"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/scale"
type: "Scale"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/relu"
type: "ReLU"
bottom: "conv8"
top: "conv8"
}
layer {
name: "conv9/dw"
type: "DepthwiseConvolution"
bottom: "conv8"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/dw/bn"
type: "BatchNorm"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/dw/scale"
type: "Scale"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/dw/relu"
type: "ReLU"
bottom: "conv9/dw"
top: "conv9/dw"
}
layer {
name: "conv9"
type: "Convolution"
bottom: "conv9/dw"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/bn"
type: "BatchNorm"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/scale"
type: "Scale"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/relu"
type: "ReLU"
bottom: "conv9"
top: "conv9"
}
layer {
name: "conv10/dw"
type: "DepthwiseConvolution"
bottom: "conv9"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/dw/bn"
type: "BatchNorm"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/dw/scale"
type: "Scale"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/dw/relu"
type: "ReLU"
bottom: "conv10/dw"
top: "conv10/dw"
}
layer {
name: "conv10"
type: "Convolution"
bottom: "conv10/dw"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/bn"
type: "BatchNorm"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/scale"
type: "Scale"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/relu"
type: "ReLU"
bottom: "conv10"
top: "conv10"
}
layer {
name: "conv11/dw"
type: "DepthwiseConvolution"
bottom: "conv10"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/dw/bn"
type: "BatchNorm"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/dw/scale"
type: "Scale"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/dw/relu"
type: "ReLU"
bottom: "conv11/dw"
top: "conv11/dw"
}
layer {
name: "conv11"
type: "Convolution"
bottom: "conv11/dw"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/bn"
type: "BatchNorm"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/scale"
type: "Scale"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/relu"
type: "ReLU"
bottom: "conv11"
top: "conv11"
}
layer {
name: "conv12/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/dw/bn"
type: "BatchNorm"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/dw/scale"
type: "Scale"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/dw/relu"
type: "ReLU"
bottom: "conv12/dw"
top: "conv12/dw"
}
layer {
name: "conv12"
type: "Convolution"
bottom: "conv12/dw"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/bn"
type: "BatchNorm"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/scale"
type: "Scale"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/relu"
type: "ReLU"
bottom: "conv12"
top: "conv12"
}
layer {
name: "conv13/dw"
type: "DepthwiseConvolution"
bottom: "conv12"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/dw/bn"
type: "BatchNorm"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/dw/scale"
type: "Scale"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/dw/relu"
type: "ReLU"
bottom: "conv13/dw"
top: "conv13/dw"
}
layer {
name: "conv13"
type: "Convolution"
bottom: "conv13/dw"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/bn"
type: "BatchNorm"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/scale"
type: "Scale"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/relu"
type: "ReLU"
bottom: "conv13"
top: "conv13"
}
layer {
name: "conv16/dw"
type: "DepthwiseConvolution"
bottom: "conv13"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/dw/bn"
type: "BatchNorm"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/dw/scale"
type: "Scale"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/dw/relu"
type: "ReLU"
bottom: "conv16/dw"
top: "conv16/dw"
}
layer {
name: "conv16"
type: "Convolution"
bottom: "conv16/dw"
top: "conv16"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/bn"
type: "BatchNorm"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/scale"
type: "Scale"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/relu"
type: "ReLU"
bottom: "conv16"
top: "conv16"
}
layer {
name: "upsample"
type: "Deconvolution"
bottom: "conv16"
top: "upsample"
param { lr_mult: 0 decay_mult: 0 }
convolution_param {
num_output: 512
kernel_size: 4 stride: 2 pad: 1
group: 512
weight_filler: { type: "bilinear" }
bias_term: false
}
}
layer {
name: "conv17/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/dw/bn"
type: "BatchNorm"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/dw/scale"
type: "Scale"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/dw/relu"
type: "ReLU"
bottom: "conv17/dw"
top: "conv17/dw"
}
layer {
name: "conv17"
type: "Convolution"
bottom: "conv17/dw"
top: "conv17"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/bn"
type: "BatchNorm"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/scale"
type: "Scale"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/relu"
type: "ReLU"
bottom: "conv17"
top: "conv17"
}
layer {
name: "conv17/sum"
type: "Eltwise"
bottom: "conv17"
bottom: "upsample"
top: "conv17/sum"
}
layer {
name: "conv18/dw"
type: "DepthwiseConvolution"
bottom: "conv17/sum"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/dw/bn"
type: "BatchNorm"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/dw/scale"
type: "Scale"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/dw/relu"
type: "ReLU"
bottom: "conv18/dw"
top: "conv18/dw"
}
layer {
name: "conv18"
type: "Convolution"
bottom: "conv18/dw"
top: "conv18"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/bn"
type: "BatchNorm"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/scale"
type: "Scale"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/relu"
type: "ReLU"
bottom: "conv18"
top: "conv18"
}
layer {
name: "conv20"
type: "Convolution"
bottom: "conv16"
top: "conv20"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "conv21"
type: "Convolution"
bottom: "conv18"
top: "conv21"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "Yolov3Loss1"
type: "Yolov3"
bottom: "conv20"
bottom: "label"
top: "det_loss1"
loss_weight: 1
yolov3_param {
side: 13
num_class: 20
num: 3
object_scale: 5.0
noobject_scale: 1.0
class_scale: 1.0
coord_scale: 1.0
thresh: 0.6
anchors_scale : 32
use_logic_gradient : false
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 10
biases: 14
biases: 23
biases: 27
biases: 37
biases: 58
biases: 81
biases: 82
biases: 135
biases: 169
biases: 344
biases: 319
mask:3
mask:4
mask:5
}
}
layer {
name: "Yolov3Loss2"
type: "Yolov3"
bottom: "conv21"
bottom: "label"
top: "det_loss2"
loss_weight: 1
yolov3_param {
side: 26
num_class: 20
num: 3
object_scale: 5.0
noobject_scale: 1.0
class_scale: 1.0
coord_scale: 1.0
thresh: 0.6
anchors_scale : 16
use_logic_gradient : false
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 10
biases: 14
biases: 23
biases: 27
biases: 37
biases: 58
biases: 81
biases: 82
biases: 135
biases: 169
biases: 344
biases: 319
mask:0
mask:1
mask:2
}
}
name: "MobileNet-YOLO"
layer {
name: "data"
type: "AnnotatedData"
top: "data"
top: "label"
include {
phase: TEST
}
transform_param {
scale: 0.007843
mean_value: 127.5
mean_value: 127.5
mean_value: 127.5
resize_param {
prob: 1.0
resize_mode: WARP
height: 416
width: 416
interp_mode: LINEAR
}
}
data_param {
source: "examples/VOC0712/VOC0712_test_lmdb"
batch_size: 1
backend: LMDB
}
annotated_data_param {
batch_sampler {
}
label_map_file: "data/VOC0712/labelmap_voc.prototxt"
}
}
layer {
name: "conv0"
type: "Convolution"
bottom: "data"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
stride: 2
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv0/bn"
type: "BatchNorm"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv0/scale"
type: "Scale"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv0/relu"
type: "ReLU"
bottom: "conv0"
top: "conv0"
}
layer {
name: "conv1/dw"
type: "DepthwiseConvolution"
bottom: "conv0"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
group: 32
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/dw/bn"
type: "BatchNorm"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/dw/scale"
type: "Scale"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/dw/relu"
type: "ReLU"
bottom: "conv1/dw"
top: "conv1/dw"
}
layer {
name: "conv1"
type: "Convolution"
bottom: "conv1/dw"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/bn"
type: "BatchNorm"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/scale"
type: "Scale"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/relu"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv2/dw"
type: "DepthwiseConvolution"
bottom: "conv1"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 64
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/dw/bn"
type: "BatchNorm"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/dw/scale"
type: "Scale"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/dw/relu"
type: "ReLU"
bottom: "conv2/dw"
top: "conv2/dw"
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv2/dw"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/bn"
type: "BatchNorm"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/scale"
type: "Scale"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/relu"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv3/dw"
type: "DepthwiseConvolution"
bottom: "conv2"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/dw/bn"
type: "BatchNorm"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/dw/scale"
type: "Scale"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/dw/relu"
type: "ReLU"
bottom: "conv3/dw"
top: "conv3/dw"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv3/dw"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/bn"
type: "BatchNorm"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/scale"
type: "Scale"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/relu"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4/dw"
type: "DepthwiseConvolution"
bottom: "conv3"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/dw/bn"
type: "BatchNorm"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/dw/scale"
type: "Scale"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/dw/relu"
type: "ReLU"
bottom: "conv4/dw"
top: "conv4/dw"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv4/dw"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/bn"
type: "BatchNorm"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/scale"
type: "Scale"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/relu"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5/dw"
type: "DepthwiseConvolution"
bottom: "conv4"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/dw/bn"
type: "BatchNorm"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/dw/scale"
type: "Scale"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/dw/relu"
type: "ReLU"
bottom: "conv5/dw"
top: "conv5/dw"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv5/dw"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/bn"
type: "BatchNorm"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/scale"
type: "Scale"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/relu"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "conv6/dw"
type: "DepthwiseConvolution"
bottom: "conv5"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/dw/bn"
type: "BatchNorm"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/dw/scale"
type: "Scale"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/dw/relu"
type: "ReLU"
bottom: "conv6/dw"
top: "conv6/dw"
}
layer {
name: "conv6"
type: "Convolution"
bottom: "conv6/dw"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/bn"
type: "BatchNorm"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/scale"
type: "Scale"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/relu"
type: "ReLU"
bottom: "conv6"
top: "conv6"
}
layer {
name: "conv7/dw"
type: "DepthwiseConvolution"
bottom: "conv6"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/dw/bn"
type: "BatchNorm"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/dw/scale"
type: "Scale"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/dw/relu"
type: "ReLU"
bottom: "conv7/dw"
top: "conv7/dw"
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv7/dw"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/bn"
type: "BatchNorm"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/scale"
type: "Scale"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/relu"
type: "ReLU"
bottom: "conv7"
top: "conv7"
}
layer {
name: "conv8/dw"
type: "DepthwiseConvolution"
bottom: "conv7"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/dw/bn"
type: "BatchNorm"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/dw/scale"
type: "Scale"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/dw/relu"
type: "ReLU"
bottom: "conv8/dw"
top: "conv8/dw"
}
layer {
name: "conv8"
type: "Convolution"
bottom: "conv8/dw"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/bn"
type: "BatchNorm"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/scale"
type: "Scale"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/relu"
type: "ReLU"
bottom: "conv8"
top: "conv8"
}
layer {
name: "conv9/dw"
type: "DepthwiseConvolution"
bottom: "conv8"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/dw/bn"
type: "BatchNorm"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/dw/scale"
type: "Scale"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/dw/relu"
type: "ReLU"
bottom: "conv9/dw"
top: "conv9/dw"
}
layer {
name: "conv9"
type: "Convolution"
bottom: "conv9/dw"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/bn"
type: "BatchNorm"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/scale"
type: "Scale"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/relu"
type: "ReLU"
bottom: "conv9"
top: "conv9"
}
layer {
name: "conv10/dw"
type: "DepthwiseConvolution"
bottom: "conv9"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/dw/bn"
type: "BatchNorm"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/dw/scale"
type: "Scale"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/dw/relu"
type: "ReLU"
bottom: "conv10/dw"
top: "conv10/dw"
}
layer {
name: "conv10"
type: "Convolution"
bottom: "conv10/dw"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/bn"
type: "BatchNorm"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/scale"
type: "Scale"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/relu"
type: "ReLU"
bottom: "conv10"
top: "conv10"
}
layer {
name: "conv11/dw"
type: "DepthwiseConvolution"
bottom: "conv10"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/dw/bn"
type: "BatchNorm"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/dw/scale"
type: "Scale"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/dw/relu"
type: "ReLU"
bottom: "conv11/dw"
top: "conv11/dw"
}
layer {
name: "conv11"
type: "Convolution"
bottom: "conv11/dw"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/bn"
type: "BatchNorm"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/scale"
type: "Scale"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/relu"
type: "ReLU"
bottom: "conv11"
top: "conv11"
}
layer {
name: "conv12/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/dw/bn"
type: "BatchNorm"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/dw/scale"
type: "Scale"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/dw/relu"
type: "ReLU"
bottom: "conv12/dw"
top: "conv12/dw"
}
layer {
name: "conv12"
type: "Convolution"
bottom: "conv12/dw"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/bn"
type: "BatchNorm"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/scale"
type: "Scale"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/relu"
type: "ReLU"
bottom: "conv12"
top: "conv12"
}
layer {
name: "conv13/dw"
type: "DepthwiseConvolution"
bottom: "conv12"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/dw/bn"
type: "BatchNorm"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/dw/scale"
type: "Scale"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/dw/relu"
type: "ReLU"
bottom: "conv13/dw"
top: "conv13/dw"
}
layer {
name: "conv13"
type: "Convolution"
bottom: "conv13/dw"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/bn"
type: "BatchNorm"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/scale"
type: "Scale"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/relu"
type: "ReLU"
bottom: "conv13"
top: "conv13"
}
layer {
name: "conv16/dw"
type: "DepthwiseConvolution"
bottom: "conv13"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/dw/bn"
type: "BatchNorm"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/dw/scale"
type: "Scale"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/dw/relu"
type: "ReLU"
bottom: "conv16/dw"
top: "conv16/dw"
}
layer {
name: "conv16"
type: "Convolution"
bottom: "conv16/dw"
top: "conv16"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/bn"
type: "BatchNorm"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/scale"
type: "Scale"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/relu"
type: "ReLU"
bottom: "conv16"
top: "conv16"
}
layer {
name: "upsample"
type: "Deconvolution"
bottom: "conv16"
top: "upsample"
param { lr_mult: 0 decay_mult: 0 }
convolution_param {
num_output: 512
kernel_size: 4 stride: 2 pad: 1
group: 512
weight_filler: { type: "bilinear" }
bias_term: false
}
}
layer {
name: "conv17/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/dw/bn"
type: "BatchNorm"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/dw/scale"
type: "Scale"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/dw/relu"
type: "ReLU"
bottom: "conv17/dw"
top: "conv17/dw"
}
layer {
name: "conv17"
type: "Convolution"
bottom: "conv17/dw"
top: "conv17"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/bn"
type: "BatchNorm"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/scale"
type: "Scale"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/relu"
type: "ReLU"
bottom: "conv17"
top: "conv17"
}
layer {
name: "conv17/sum"
type: "Eltwise"
bottom: "conv17"
bottom: "upsample"
top: "conv17/sum"
}
layer {
name: "conv18/dw"
type: "DepthwiseConvolution"
bottom: "conv17/sum"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/dw/bn"
type: "BatchNorm"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/dw/scale"
type: "Scale"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/dw/relu"
type: "ReLU"
bottom: "conv18/dw"
top: "conv18/dw"
}
layer {
name: "conv18"
type: "Convolution"
bottom: "conv18/dw"
top: "conv18"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/bn"
type: "BatchNorm"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/scale"
type: "Scale"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/relu"
type: "ReLU"
bottom: "conv18"
top: "conv18"
}
layer {
name: "conv20"
type: "Convolution"
bottom: "conv16"
top: "conv20"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "conv21"
type: "Convolution"
bottom: "conv18"
top: "conv21"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "detection_out"
type: "Yolov3DetectionOutput"
bottom: "conv20"
bottom: "conv21"
top: "detection_out"
include {
phase: TEST
}
yolov3_detection_output_param {
confidence_threshold: 0.01
nms_threshold: 0.55
num_classes: 20
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 10
biases: 14
biases: 23
biases: 27
biases: 37
biases: 58
biases: 81
biases: 82
biases: 135
biases: 169
biases: 344
biases: 319
mask:3
mask:4
mask:5
mask:0
mask:1
mask:2
anchors_scale:32
anchors_scale:16
mask_group_num:2
}
}
layer {
name: "detection_eval"
type: "DetectionEvaluate"
bottom: "detection_out"
bottom: "label"
top: "detection_eval"
include {
phase: TEST
}
detection_evaluate_param {
num_classes: 21
background_label_id: 0
overlap_threshold: 0.5
evaluate_difficult_gt: false
}
}
here ,i am not using pre-caffemodel to train ,just
argc=3
argv[1]="train"
,argv[2]="--solver=***.prototxt "
when i test the train_model it is runing ok ,but not get good result ,here not merge bn in deploy ,and not change the model weights . it is some error in this way .
following is deploy file ;
name: "MobileNet-YOLO"
layer {
name: "data"
type: "Input"
top: "data"
input_param {
shape {
dim: 1
dim: 3
dim: 320
dim: 320
}
}
}
layer {
name: "conv0"
type: "Convolution"
bottom: "data"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
stride: 2
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv0/bn"
type: "BatchNorm"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv0/scale"
type: "Scale"
bottom: "conv0"
top: "conv0"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv0/relu"
type: "ReLU"
bottom: "conv0"
top: "conv0"
}
layer {
name: "conv1/dw"
type: "DepthwiseConvolution"
bottom: "conv0"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 32
bias_term: false
pad: 1
kernel_size: 3
group: 32
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/dw/bn"
type: "BatchNorm"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/dw/scale"
type: "Scale"
bottom: "conv1/dw"
top: "conv1/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/dw/relu"
type: "ReLU"
bottom: "conv1/dw"
top: "conv1/dw"
}
layer {
name: "conv1"
type: "Convolution"
bottom: "conv1/dw"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv1/bn"
type: "BatchNorm"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv1/scale"
type: "Scale"
bottom: "conv1"
top: "conv1"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv1/relu"
type: "ReLU"
bottom: "conv1"
top: "conv1"
}
layer {
name: "conv2/dw"
type: "DepthwiseConvolution"
bottom: "conv1"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 64
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 64
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/dw/bn"
type: "BatchNorm"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/dw/scale"
type: "Scale"
bottom: "conv2/dw"
top: "conv2/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/dw/relu"
type: "ReLU"
bottom: "conv2/dw"
top: "conv2/dw"
}
layer {
name: "conv2"
type: "Convolution"
bottom: "conv2/dw"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv2/bn"
type: "BatchNorm"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv2/scale"
type: "Scale"
bottom: "conv2"
top: "conv2"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv2/relu"
type: "ReLU"
bottom: "conv2"
top: "conv2"
}
layer {
name: "conv3/dw"
type: "DepthwiseConvolution"
bottom: "conv2"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/dw/bn"
type: "BatchNorm"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/dw/scale"
type: "Scale"
bottom: "conv3/dw"
top: "conv3/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/dw/relu"
type: "ReLU"
bottom: "conv3/dw"
top: "conv3/dw"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "conv3/dw"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv3/bn"
type: "BatchNorm"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv3/scale"
type: "Scale"
bottom: "conv3"
top: "conv3"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv3/relu"
type: "ReLU"
bottom: "conv3"
top: "conv3"
}
layer {
name: "conv4/dw"
type: "DepthwiseConvolution"
bottom: "conv3"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 128
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 128
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/dw/bn"
type: "BatchNorm"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/dw/scale"
type: "Scale"
bottom: "conv4/dw"
top: "conv4/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/dw/relu"
type: "ReLU"
bottom: "conv4/dw"
top: "conv4/dw"
}
layer {
name: "conv4"
type: "Convolution"
bottom: "conv4/dw"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv4/bn"
type: "BatchNorm"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv4/scale"
type: "Scale"
bottom: "conv4"
top: "conv4"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv4/relu"
type: "ReLU"
bottom: "conv4"
top: "conv4"
}
layer {
name: "conv5/dw"
type: "DepthwiseConvolution"
bottom: "conv4"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/dw/bn"
type: "BatchNorm"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/dw/scale"
type: "Scale"
bottom: "conv5/dw"
top: "conv5/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/dw/relu"
type: "ReLU"
bottom: "conv5/dw"
top: "conv5/dw"
}
layer {
name: "conv5"
type: "Convolution"
bottom: "conv5/dw"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv5/bn"
type: "BatchNorm"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv5/scale"
type: "Scale"
bottom: "conv5"
top: "conv5"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv5/relu"
type: "ReLU"
bottom: "conv5"
top: "conv5"
}
layer {
name: "conv6/dw"
type: "DepthwiseConvolution"
bottom: "conv5"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 256
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 256
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/dw/bn"
type: "BatchNorm"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/dw/scale"
type: "Scale"
bottom: "conv6/dw"
top: "conv6/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/dw/relu"
type: "ReLU"
bottom: "conv6/dw"
top: "conv6/dw"
}
layer {
name: "conv6"
type: "Convolution"
bottom: "conv6/dw"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv6/bn"
type: "BatchNorm"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv6/scale"
type: "Scale"
bottom: "conv6"
top: "conv6"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv6/relu"
type: "ReLU"
bottom: "conv6"
top: "conv6"
}
layer {
name: "conv7/dw"
type: "DepthwiseConvolution"
bottom: "conv6"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/dw/bn"
type: "BatchNorm"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/dw/scale"
type: "Scale"
bottom: "conv7/dw"
top: "conv7/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/dw/relu"
type: "ReLU"
bottom: "conv7/dw"
top: "conv7/dw"
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv7/dw"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv7/bn"
type: "BatchNorm"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv7/scale"
type: "Scale"
bottom: "conv7"
top: "conv7"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv7/relu"
type: "ReLU"
bottom: "conv7"
top: "conv7"
}
layer {
name: "conv8/dw"
type: "DepthwiseConvolution"
bottom: "conv7"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/dw/bn"
type: "BatchNorm"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/dw/scale"
type: "Scale"
bottom: "conv8/dw"
top: "conv8/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/dw/relu"
type: "ReLU"
bottom: "conv8/dw"
top: "conv8/dw"
}
layer {
name: "conv8"
type: "Convolution"
bottom: "conv8/dw"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv8/bn"
type: "BatchNorm"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv8/scale"
type: "Scale"
bottom: "conv8"
top: "conv8"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv8/relu"
type: "ReLU"
bottom: "conv8"
top: "conv8"
}
layer {
name: "conv9/dw"
type: "DepthwiseConvolution"
bottom: "conv8"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/dw/bn"
type: "BatchNorm"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/dw/scale"
type: "Scale"
bottom: "conv9/dw"
top: "conv9/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/dw/relu"
type: "ReLU"
bottom: "conv9/dw"
top: "conv9/dw"
}
layer {
name: "conv9"
type: "Convolution"
bottom: "conv9/dw"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv9/bn"
type: "BatchNorm"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv9/scale"
type: "Scale"
bottom: "conv9"
top: "conv9"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv9/relu"
type: "ReLU"
bottom: "conv9"
top: "conv9"
}
layer {
name: "conv10/dw"
type: "DepthwiseConvolution"
bottom: "conv9"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/dw/bn"
type: "BatchNorm"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/dw/scale"
type: "Scale"
bottom: "conv10/dw"
top: "conv10/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/dw/relu"
type: "ReLU"
bottom: "conv10/dw"
top: "conv10/dw"
}
layer {
name: "conv10"
type: "Convolution"
bottom: "conv10/dw"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv10/bn"
type: "BatchNorm"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv10/scale"
type: "Scale"
bottom: "conv10"
top: "conv10"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv10/relu"
type: "ReLU"
bottom: "conv10"
top: "conv10"
}
layer {
name: "conv11/dw"
type: "DepthwiseConvolution"
bottom: "conv10"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/dw/bn"
type: "BatchNorm"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/dw/scale"
type: "Scale"
bottom: "conv11/dw"
top: "conv11/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/dw/relu"
type: "ReLU"
bottom: "conv11/dw"
top: "conv11/dw"
}
layer {
name: "conv11"
type: "Convolution"
bottom: "conv11/dw"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv11/bn"
type: "BatchNorm"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv11/scale"
type: "Scale"
bottom: "conv11"
top: "conv11"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv11/relu"
type: "ReLU"
bottom: "conv11"
top: "conv11"
}
layer {
name: "conv12/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
stride: 2
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/dw/bn"
type: "BatchNorm"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/dw/scale"
type: "Scale"
bottom: "conv12/dw"
top: "conv12/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/dw/relu"
type: "ReLU"
bottom: "conv12/dw"
top: "conv12/dw"
}
layer {
name: "conv12"
type: "Convolution"
bottom: "conv12/dw"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv12/bn"
type: "BatchNorm"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv12/scale"
type: "Scale"
bottom: "conv12"
top: "conv12"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv12/relu"
type: "ReLU"
bottom: "conv12"
top: "conv12"
}
layer {
name: "conv13/dw"
type: "DepthwiseConvolution"
bottom: "conv12"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/dw/bn"
type: "BatchNorm"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/dw/scale"
type: "Scale"
bottom: "conv13/dw"
top: "conv13/dw"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/dw/relu"
type: "ReLU"
bottom: "conv13/dw"
top: "conv13/dw"
}
layer {
name: "conv13"
type: "Convolution"
bottom: "conv13/dw"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv13/bn"
type: "BatchNorm"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv13/scale"
type: "Scale"
bottom: "conv13"
top: "conv13"
param {
lr_mult: 0.1
decay_mult: 0.0
}
param {
lr_mult: 0.2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv13/relu"
type: "ReLU"
bottom: "conv13"
top: "conv13"
}
layer {
name: "conv16/dw"
type: "DepthwiseConvolution"
bottom: "conv13"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
pad: 1
kernel_size: 3
group: 1024
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/dw/bn"
type: "BatchNorm"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/dw/scale"
type: "Scale"
bottom: "conv16/dw"
top: "conv16/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/dw/relu"
type: "ReLU"
bottom: "conv16/dw"
top: "conv16/dw"
}
layer {
name: "conv16"
type: "Convolution"
bottom: "conv16/dw"
top: "conv16"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv16/bn"
type: "BatchNorm"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv16/scale"
type: "Scale"
bottom: "conv16"
top: "conv16"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv16/relu"
type: "ReLU"
bottom: "conv16"
top: "conv16"
}
layer {
name: "upsample"
type: "Deconvolution"
bottom: "conv16"
top: "upsample"
param { lr_mult: 0 decay_mult: 0 }
convolution_param {
num_output: 512
kernel_size: 4 stride: 2 pad: 1
group: 512
weight_filler: { type: "bilinear" }
bias_term: false
}
}
layer {
name: "conv17/dw"
type: "DepthwiseConvolution"
bottom: "conv11"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/dw/bn"
type: "BatchNorm"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/dw/scale"
type: "Scale"
bottom: "conv17/dw"
top: "conv17/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/dw/relu"
type: "ReLU"
bottom: "conv17/dw"
top: "conv17/dw"
}
layer {
name: "conv17"
type: "Convolution"
bottom: "conv17/dw"
top: "conv17"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv17/bn"
type: "BatchNorm"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv17/scale"
type: "Scale"
bottom: "conv17"
top: "conv17"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv17/relu"
type: "ReLU"
bottom: "conv17"
top: "conv17"
}
layer {
name: "conv17/sum"
type: "Eltwise"
bottom: "conv17"
bottom: "upsample"
top: "conv17/sum"
}
layer {
name: "conv18/dw"
type: "DepthwiseConvolution"
bottom: "conv17/sum"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 512
bias_term: false
pad: 1
kernel_size: 3
group: 512
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/dw/bn"
type: "BatchNorm"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/dw/scale"
type: "Scale"
bottom: "conv18/dw"
top: "conv18/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/dw/relu"
type: "ReLU"
bottom: "conv18/dw"
top: "conv18/dw"
}
layer {
name: "conv18"
type: "Convolution"
bottom: "conv18/dw"
top: "conv18"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 1024
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "conv18/bn"
type: "BatchNorm"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
param {
lr_mult: 0
decay_mult: 0
}
}
layer {
name: "conv18/scale"
type: "Scale"
bottom: "conv18"
top: "conv18"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "conv18/relu"
type: "ReLU"
bottom: "conv18"
top: "conv18"
}
layer {
name: "conv20"
type: "Convolution"
bottom: "conv16"
top: "conv20"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "conv21"
type: "Convolution"
bottom: "conv18"
top: "conv21"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 75
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "detection_out"
type: "Yolov3DetectionOutput"
bottom: "conv20"
bottom: "conv21"
top: "detection_out"
include {
phase: TEST
}
yolov3_detection_output_param {
confidence_threshold: 0.01
nms_threshold: 0.45
num_classes: 20
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 10
biases: 14
biases: 23
biases: 27
biases: 37
biases: 58
biases: 81
biases: 82
biases: 135
biases: 169
biases: 344
biases: 319
mask:3
mask:4
mask:5
mask:0
mask:1
mask:2
anchors_scale:32
anchors_scale:16
mask_group_num:2
}
}
thankes !!
About yolov3 have problem!
Hi ！First of all, thank you for your contribution to the open source community.
My English is very poor ,but i will tyy my best to Describe my problem。
I got that files through Caffe-YOLOv3-Windows.
in this mobilenet_yolov3_lite is very great!
About yolov3 I shoud to ncnn (and not mobilenet_yolov3 or mobilenet_yolov3_lite) .
I get the caffemodel in Oringinal darknet-yolov3 links ,But because of the upsample problem, I need to convert it to ncnn->interp, the final result is too different, I read the code inside ncnn interp.cpp, I don't know what the problem is, so I want to ask if it is yolov3— Caffemodel has been iterated 100 times without much training. With same yolov3-spp .
the last one about mobilenet_yolov3 . I can get the Caffemodel for mobilenet_yolov3 ,but can't
get the deploy .prototxt . to link graph I can only get a picture description of the network structure through netscope, but can't find mobilenet_yolov3_deploy.prototxt.
memory leak
memory leak
你好我有两个问题
1、你这又yolov3的测试脚本吗，不是yolov3 lite或者mobilenetyolov3？
2、我该怎么修改才能用自己的yolov3的caffemodel测试自己的数据？
The convert tools doesn't seem to work
The convert tools doesn't seem to work.
$ python create_yolo_prototxt.py yolov2.cfg yolov2.prototxt
ERROR:root:route layer is not supported
ERROR:root:reorg layer is not supported
ERROR:root:route layer is not supported
ERROR:root:region layer is not supported
Proactively let negative samples participate in training problems(主动让负样本参与训练问题)
Hi guys,
In engineering projects, there are often situations in which a picture is completely unmarked. In the process of detecting, there are many misidentified results.
Previously, in the darknet framework, directly adding some unlabeled information to the training data set can solve these problems.
Now, I want to use caffe to train this model of yolov3, but I find it seems that I can't actively add images with only negative samples for training.
If you want to train by directly adding pure negative samples, how to deal with it, thank you~
在实地的工程项目中，经常会遇到一些情况，就是一个完全没有标记样本的图片，在进行检测的过程中，有很多被误识别的结果出来。
之前在darknet框架下的方式是，直接增加一些没有标注的信息的图片，增加到训练数据集中，可以去解决这些问题。
现在，我想用caffe去训练一下yolov3的这个模型，但是发现好像不能去主动添加完全只有负样本的图片进行训练。
如果想通过直接增加纯负样本的方式来进行训练，该怎么处理，谢谢~
@eric612 大佬 能否指点一下 谢谢
训练2类目标
你好，请问用yolov3_lite训练两类目标的时候，网络配置文件该如何修改呢？
我改了conv20,conv21的num_output为18
Yolov3Loss1，Yolov3Loss2的num_class为2
训练报错：
yolov3_layer.cpp:143] Check failed: input_count == tmp_input_count (6498 vs. 7581)
求正确的改法~
some different between darknet with your yolov3?
in darknet
- obj/no obj loss
l.delta[obj_index] = 0 - l.output[obj_index]; // no obj diff
if (best_iou > l.ignore_thresh) {
l.delta[obj_index] = 0;
}
if (best_iou > l.truth_thresh) {
l.delta[obj_index] = 1 - l.output[obj_index]; //obj diff
}
why your yolov3 has no ignore_thresh ? - you swap data through logistic_activate
caffemodel for yolov3 (windows)
Hi Eric, are you able to upload the caffemodel for yolo v3 (windows)? In your link below, you have only uploaded the yolov3 lite model.
https://github.com/eric612/Caffe-YOLOv3-Windows/tree/master/models/yolov3
MobileNetYolo_deploy.caffemodel
Hi, Eric.
It's me, Daniel. Again :)
Would you pls update the MobileNetYolo_deploy.caffemodel file to github?
Or is it hidden in somewhere that I can't find it.
Cheers
训练mobilenet_yolov3模型，如果想训练320X320，需要修改哪些地方？
我修改了side的值，训练报错。
map 57.9 on voc2007test
I test voc2007 with mobilenet_yolov3_lite_deploy
but the map has only 57.9
I don't know why?
检测输出为[0,-1,-1,-1,-1,-1,-1,-1]
您好，我在windows下使用detection_out.cpp推理时，用models\yolov3路径下的模型测试检测正常，但是切换到自己的模型测试时检测层输出一直为[0,-1,-1,-1,-1,-1,-1,-1]，模型是使用您linux版本的环境训练的，训练时map值正常。我将caffemodel转成ncnn，在ncnn下测试也正常。我怀疑可能是deploy文件中layer层与linux环境中的layer不一样导致，但是比对只后未发现异常，这个问题困扰我好多天了，请求您的帮助！这是我的deploy文件。
name: "yolov3"
input: "data"
layer {
name: "data"
type: "Input"
top: "data"
input_param {shape {dim: 1 dim: 3 dim: 800 dim: 800 } }
}
layer {
name: "conv1"
type: "Convolution"
bottom: "data"
top: "conv1"
convolution_param {
num_output: 16
bias_term: false
pad: 1
kernel_size: 3
group: 1
stride: 2
weight_filler {
type: "msra"
}
dilation: 1
}
}
layer {
name: "bn_scale1"
type: "Scale"
bottom: "conv1"
top: "conv1"
scale_param {
bias_term: true
}
}
layer {
name: "power1"
type: "Power"
bottom: "conv1"
top: "power1"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus1"
type: "ReLU6"
bottom: "power1"
top: "relus1"
}
layer {
name: "power2"
type: "Power"
bottom: "relus1"
top: "power2"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul1"
type: "Eltwise"
bottom: "conv1"
bottom: "power2"
top: "mul1"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv2"
type: "DepthwiseConvolution"
bottom: "mul1"
top: "conv2"
convolution_param {
num_output: 16
bias_term: false
pad: 1
kernel_size: 3
group: 16
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale2"
type: "Scale"
bottom: "conv2"
top: "conv2"
scale_param {
bias_term: true
}
}
layer {
name: "relu1"
type: "ReLU"
bottom: "conv2"
top: "relu1"
}
layer {
name: "conv3"
type: "Convolution"
bottom: "relu1"
top: "conv3"
convolution_param {
num_output: 16
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale3"
type: "Scale"
bottom: "conv3"
top: "conv3"
scale_param {
bias_term: true
}
}
layer {
name: "add1"
type: "Eltwise"
bottom: "mul1"
bottom: "conv3"
top: "add1"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv4"
type: "Convolution"
bottom: "add1"
top: "conv4"
convolution_param {
num_output: 64
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale4"
type: "Scale"
bottom: "conv4"
top: "conv4"
scale_param {
bias_term: true
}
}
layer {
name: "relu2"
type: "ReLU"
bottom: "conv4"
top: "relu2"
}
layer {
name: "conv5"
type: "DepthwiseConvolution"
bottom: "relu2"
top: "conv5"
convolution_param {
num_output: 64
bias_term: false
pad: 1
kernel_size: 3
group: 64
stride: 2
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale5"
type: "Scale"
bottom: "conv5"
top: "conv5"
scale_param {
bias_term: true
}
}
layer {
name: "relu3"
type: "ReLU"
bottom: "conv5"
top: "relu3"
}
layer {
name: "conv6"
type: "Convolution"
bottom: "relu3"
top: "conv6"
convolution_param {
num_output: 24
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale6"
type: "Scale"
bottom: "conv6"
top: "conv6"
scale_param {
bias_term: true
}
}
layer {
name: "conv7"
type: "Convolution"
bottom: "conv6"
top: "conv7"
convolution_param {
num_output: 72
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale7"
type: "Scale"
bottom: "conv7"
top: "conv7"
scale_param {
bias_term: true
}
}
layer {
name: "relu4"
type: "ReLU"
bottom: "conv7"
top: "relu4"
}
layer {
name: "conv8"
type: "DepthwiseConvolution"
bottom: "relu4"
top: "conv8"
convolution_param {
num_output: 72
bias_term: false
pad: 1
kernel_size: 3
group: 72
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale8"
type: "Scale"
bottom: "conv8"
top: "conv8"
scale_param {
bias_term: true
}
}
layer {
name: "relu5"
type: "ReLU"
bottom: "conv8"
top: "relu5"
}
layer {
name: "conv9"
type: "Convolution"
bottom: "relu5"
top: "conv9"
convolution_param {
num_output: 24
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale9"
type: "Scale"
bottom: "conv9"
top: "conv9"
scale_param {
bias_term: true
}
}
layer {
name: "add2"
type: "Eltwise"
bottom: "conv6"
bottom: "conv9"
top: "add2"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv10"
type: "Convolution"
bottom: "add2"
top: "conv10"
convolution_param {
num_output: 72
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale10"
type: "Scale"
bottom: "conv10"
top: "conv10"
scale_param {
bias_term: true
}
}
layer {
name: "relu6"
type: "ReLU"
bottom: "conv10"
top: "relu6"
}
layer {
name: "conv11"
type: "DepthwiseConvolution"
bottom: "relu6"
top: "conv11"
convolution_param {
num_output: 72
bias_term: false
pad: 2
kernel_size: 5
group: 72
stride: 2
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "ave_pool1"
type: "Pooling"
bottom: "conv11"
top: "ave_pool1"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route1"
type: "Concat"
bottom: "ave_pool1"
top: "route1"
}
layer {
name: "fc1"
type: "InnerProduct"
bottom: "route1"
top: "fc1"
inner_product_param {
num_output: 18
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu7"
type: "ReLU"
bottom: "fc1"
top: "relu7"
}
layer {
name: "fc2"
type: "InnerProduct"
bottom: "relu7"
top: "fc2"
inner_product_param {
num_output: 72
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power3"
type: "Power"
bottom: "fc2"
top: "power3"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus2"
type: "ReLU6"
bottom: "power3"
top: "relus2"
}
layer {
name: "power4"
type: "Power"
bottom: "relus2"
top: "power4"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route2"
type: "Concat"
bottom: "power4"
top: "route2"
}
layer {
name: "scale1"
type: "Scale"
bottom: "conv11"
bottom: "route2"
top: "scale1"
scale_param {
axis: 0
}
}
layer {
name: "relu8"
type: "ReLU"
bottom: "scale1"
top: "relu8"
}
layer {
name: "conv12"
type: "Convolution"
bottom: "relu8"
top: "conv12"
convolution_param {
num_output: 40
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale12"
type: "Scale"
bottom: "conv12"
top: "conv12"
scale_param {
bias_term: true
}
}
layer {
name: "conv13"
type: "Convolution"
bottom: "conv12"
top: "conv13"
convolution_param {
num_output: 120
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale13"
type: "Scale"
bottom: "conv13"
top: "conv13"
scale_param {
bias_term: true
}
}
layer {
name: "relu9"
type: "ReLU"
bottom: "conv13"
top: "relu9"
}
layer {
name: "conv14"
type: "DepthwiseConvolution"
bottom: "relu9"
top: "conv14"
convolution_param {
num_output: 120
bias_term: false
pad: 2
kernel_size: 5
group: 120
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale14"
type: "Scale"
bottom: "conv14"
top: "conv14"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool2"
type: "Pooling"
bottom: "conv14"
top: "ave_pool2"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route3"
type: "Concat"
bottom: "ave_pool2"
top: "route3"
}
layer {
name: "fc3"
type: "InnerProduct"
bottom: "route3"
top: "fc3"
inner_product_param {
num_output: 30
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu10"
type: "ReLU"
bottom: "fc3"
top: "relu10"
}
layer {
name: "fc4"
type: "InnerProduct"
bottom: "relu10"
top: "fc4"
inner_product_param {
num_output: 120
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power5"
type: "Power"
bottom: "fc4"
top: "power5"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus3"
type: "ReLU6"
bottom: "power5"
top: "relus3"
}
layer {
name: "power6"
type: "Power"
bottom: "relus3"
top: "power6"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route4"
type: "Concat"
bottom: "power6"
top: "route4"
}
layer {
name: "scale2"
type: "Scale"
bottom: "conv14"
bottom: "route4"
top: "scale2"
scale_param {
axis: 0
}
}
layer {
name: "relu11"
type: "ReLU"
bottom: "scale2"
top: "relu11"
}
layer {
name: "conv15"
type: "Convolution"
bottom: "relu11"
top: "conv15"
convolution_param {
num_output: 40
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale15"
type: "Scale"
bottom: "conv15"
top: "conv15"
scale_param {
bias_term: true
}
}
layer {
name: "add3"
type: "Eltwise"
bottom: "conv12"
bottom: "conv15"
top: "add3"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv16"
type: "Convolution"
bottom: "add3"
top: "conv16"
convolution_param {
num_output: 120
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale16"
type: "Scale"
bottom: "conv16"
top: "conv16"
scale_param {
bias_term: true
}
}
layer {
name: "relu12"
type: "ReLU"
bottom: "conv16"
top: "relu12"
}
layer {
name: "conv17"
type: "DepthwiseConvolution"
bottom: "relu12"
top: "conv17"
convolution_param {
num_output: 120
bias_term: false
pad: 2
kernel_size: 5
group: 120
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale17"
type: "Scale"
bottom: "conv17"
top: "conv17"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool3"
type: "Pooling"
bottom: "conv17"
top: "ave_pool3"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route5"
type: "Concat"
bottom: "ave_pool3"
top: "route5"
}
layer {
name: "fc5"
type: "InnerProduct"
bottom: "route5"
top: "fc5"
inner_product_param {
num_output: 30
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu13"
type: "ReLU"
bottom: "fc5"
top: "relu13"
}
layer {
name: "fc6"
type: "InnerProduct"
bottom: "relu13"
top: "fc6"
inner_product_param {
num_output: 120
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power7"
type: "Power"
bottom: "fc6"
top: "power7"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus4"
type: "ReLU6"
bottom: "power7"
top: "relus4"
}
layer {
name: "power8"
type: "Power"
bottom: "relus4"
top: "power8"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route6"
type: "Concat"
bottom: "power8"
top: "route6"
}
layer {
name: "scale3"
type: "Scale"
bottom: "conv17"
bottom: "route6"
top: "scale3"
scale_param {
axis: 0
}
}
layer {
name: "relu14"
type: "ReLU"
bottom: "scale3"
top: "relu14"
}
layer {
name: "conv18"
type: "Convolution"
bottom: "relu14"
top: "conv18"
convolution_param {
num_output: 40
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale18"
type: "Scale"
bottom: "conv18"
top: "conv18"
scale_param {
bias_term: true
}
}
layer {
name: "add4"
type: "Eltwise"
bottom: "add3"
bottom: "conv18"
top: "add4"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv19"
type: "Convolution"
bottom: "add4"
top: "conv19"
convolution_param {
num_output: 240
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale19"
type: "Scale"
bottom: "conv19"
top: "conv19"
scale_param {
bias_term: true
}
}
layer {
name: "power9"
type: "Power"
bottom: "conv19"
top: "power9"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus5"
type: "ReLU6"
bottom: "power9"
top: "relus5"
}
layer {
name: "power10"
type: "Power"
bottom: "relus5"
top: "power10"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul2"
type: "Eltwise"
bottom: "conv19"
bottom: "power10"
top: "mul2"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv20"
type: "DepthwiseConvolution"
bottom: "mul2"
top: "conv20"
convolution_param {
num_output: 240
bias_term: false
pad: 1
kernel_size: 3
group: 240
stride: 2
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale20"
type: "Scale"
bottom: "conv20"
top: "conv20"
scale_param {
bias_term: true
}
}
layer {
name: "power11"
type: "Power"
bottom: "conv20"
top: "power11"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus6"
type: "ReLU6"
bottom: "power11"
top: "relus6"
}
layer {
name: "power12"
type: "Power"
bottom: "relus6"
top: "power12"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul3"
type: "Eltwise"
bottom: "conv20"
bottom: "power12"
top: "mul3"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv21"
type: "Convolution"
bottom: "mul3"
top: "conv21"
convolution_param {
num_output: 80
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale21"
type: "Scale"
bottom: "conv21"
top: "conv21"
scale_param {
bias_term: true
}
}
layer {
name: "conv22"
type: "Convolution"
bottom: "conv21"
top: "conv22"
convolution_param {
num_output: 200
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale22"
type: "Scale"
bottom: "conv22"
top: "conv22"
scale_param {
bias_term: true
}
}
layer {
name: "power13"
type: "Power"
bottom: "conv22"
top: "power13"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus7"
type: "ReLU6"
bottom: "power13"
top: "relus7"
}
layer {
name: "power14"
type: "Power"
bottom: "relus7"
top: "power14"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul4"
type: "Eltwise"
bottom: "conv22"
bottom: "power14"
top: "mul4"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv23"
type: "DepthwiseConvolution"
bottom: "mul4"
top: "conv23"
convolution_param {
num_output: 200
bias_term: false
pad: 1
kernel_size: 3
group: 200
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale23"
type: "Scale"
bottom: "conv23"
top: "conv23"
scale_param {
bias_term: true
}
}
layer {
name: "power15"
type: "Power"
bottom: "conv23"
top: "power15"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus8"
type: "ReLU6"
bottom: "power15"
top: "relus8"
}
layer {
name: "power16"
type: "Power"
bottom: "relus8"
top: "power16"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul5"
type: "Eltwise"
bottom: "conv23"
bottom: "power16"
top: "mul5"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv24"
type: "Convolution"
bottom: "mul5"
top: "conv24"
convolution_param {
num_output: 80
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale24"
type: "Scale"
bottom: "conv24"
top: "conv24"
scale_param {
bias_term: true
}
}
layer {
name: "add5"
type: "Eltwise"
bottom: "conv21"
bottom: "conv24"
top: "add5"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv25"
type: "Convolution"
bottom: "add5"
top: "conv25"
convolution_param {
num_output: 184
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale25"
type: "Scale"
bottom: "conv25"
top: "conv25"
scale_param {
bias_term: true
}
}
layer {
name: "power17"
type: "Power"
bottom: "conv25"
top: "power17"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus9"
type: "ReLU6"
bottom: "power17"
top: "relus9"
}
layer {
name: "power18"
type: "Power"
bottom: "relus9"
top: "power18"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul6"
type: "Eltwise"
bottom: "conv25"
bottom: "power18"
top: "mul6"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv26"
type: "DepthwiseConvolution"
bottom: "mul6"
top: "conv26"
convolution_param {
num_output: 184
bias_term: false
pad: 1
kernel_size: 3
group: 184
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale26"
type: "Scale"
bottom: "conv26"
top: "conv26"
scale_param {
bias_term: true
}
}
layer {
name: "power19"
type: "Power"
bottom: "conv26"
top: "power19"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus10"
type: "ReLU6"
bottom: "power19"
top: "relus10"
}
layer {
name: "power20"
type: "Power"
bottom: "relus10"
top: "power20"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul7"
type: "Eltwise"
bottom: "conv26"
bottom: "power20"
top: "mul7"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv27"
type: "Convolution"
bottom: "mul7"
top: "conv27"
convolution_param {
num_output: 80
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale27"
type: "Scale"
bottom: "conv27"
top: "conv27"
scale_param {
bias_term: true
}
}
layer {
name: "add6"
type: "Eltwise"
bottom: "add5"
bottom: "conv27"
top: "add6"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv28"
type: "Convolution"
bottom: "add6"
top: "conv28"
convolution_param {
num_output: 184
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale28"
type: "Scale"
bottom: "conv28"
top: "conv28"
scale_param {
bias_term: true
}
}
layer {
name: "power21"
type: "Power"
bottom: "conv28"
top: "power21"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus11"
type: "ReLU6"
bottom: "power21"
top: "relus11"
}
layer {
name: "power22"
type: "Power"
bottom: "relus11"
top: "power22"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul8"
type: "Eltwise"
bottom: "conv28"
bottom: "power22"
top: "mul8"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv29"
type: "DepthwiseConvolution"
bottom: "mul8"
top: "conv29"
convolution_param {
num_output: 184
bias_term: false
pad: 1
kernel_size: 3
group: 184
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale29"
type: "Scale"
bottom: "conv29"
top: "conv29"
scale_param {
bias_term: true
}
}
layer {
name: "power23"
type: "Power"
bottom: "conv29"
top: "power23"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus12"
type: "ReLU6"
bottom: "power23"
top: "relus12"
}
layer {
name: "power24"
type: "Power"
bottom: "relus12"
top: "power24"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul9"
type: "Eltwise"
bottom: "conv29"
bottom: "power24"
top: "mul9"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv30"
type: "Convolution"
bottom: "mul9"
top: "conv30"
convolution_param {
num_output: 80
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale30"
type: "Scale"
bottom: "conv30"
top: "conv30"
scale_param {
bias_term: true
}
}
layer {
name: "add7"
type: "Eltwise"
bottom: "add6"
bottom: "conv30"
top: "add7"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv31"
type: "Convolution"
bottom: "add7"
top: "conv31"
convolution_param {
num_output: 480
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale31"
type: "Scale"
bottom: "conv31"
top: "conv31"
scale_param {
bias_term: true
}
}
layer {
name: "power25"
type: "Power"
bottom: "conv31"
top: "power25"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus13"
type: "ReLU6"
bottom: "power25"
top: "relus13"
}
layer {
name: "power26"
type: "Power"
bottom: "relus13"
top: "power26"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul10"
type: "Eltwise"
bottom: "conv31"
bottom: "power26"
top: "mul10"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv32"
type: "DepthwiseConvolution"
bottom: "mul10"
top: "conv32"
convolution_param {
num_output: 480
bias_term: false
pad: 1
kernel_size: 3
group: 480
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale32"
type: "Scale"
bottom: "conv32"
top: "conv32"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool4"
type: "Pooling"
bottom: "conv32"
top: "ave_pool4"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route7"
type: "Concat"
bottom: "ave_pool4"
top: "route7"
}
layer {
name: "fc7"
type: "InnerProduct"
bottom: "route7"
top: "fc7"
inner_product_param {
num_output: 120
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu15"
type: "ReLU"
bottom: "fc7"
top: "relu15"
}
layer {
name: "fc8"
type: "InnerProduct"
bottom: "relu15"
top: "fc8"
inner_product_param {
num_output: 480
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power27"
type: "Power"
bottom: "fc8"
top: "power27"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus14"
type: "ReLU6"
bottom: "power27"
top: "relus14"
}
layer {
name: "power28"
type: "Power"
bottom: "relus14"
top: "power28"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route8"
type: "Concat"
bottom: "power28"
top: "route8"
}
layer {
name: "scale4"
type: "Scale"
bottom: "conv32"
bottom: "route8"
top: "scale4"
scale_param {
axis: 0
}
}
layer {
name: "power29"
type: "Power"
bottom: "scale4"
top: "power29"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus15"
type: "ReLU6"
bottom: "power29"
top: "relus15"
}
layer {
name: "power30"
type: "Power"
bottom: "relus15"
top: "power30"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul11"
type: "Eltwise"
bottom: "scale4"
bottom: "power30"
top: "mul11"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv33"
type: "Convolution"
bottom: "mul11"
top: "conv33"
convolution_param {
num_output: 112
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale33"
type: "Scale"
bottom: "conv33"
top: "conv33"
scale_param {
bias_term: true
}
}
layer {
name: "conv34"
type: "Convolution"
bottom: "conv33"
top: "conv34"
convolution_param {
num_output: 672
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale34"
type: "Scale"
bottom: "conv34"
top: "conv34"
scale_param {
bias_term: true
}
}
layer {
name: "power31"
type: "Power"
bottom: "conv34"
top: "power31"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus16"
type: "ReLU6"
bottom: "power31"
top: "relus16"
}
layer {
name: "power32"
type: "Power"
bottom: "relus16"
top: "power32"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul12"
type: "Eltwise"
bottom: "conv34"
bottom: "power32"
top: "mul12"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv35"
type: "DepthwiseConvolution"
bottom: "mul12"
top: "conv35"
convolution_param {
num_output: 672
bias_term: false
pad: 1
kernel_size: 3
group: 672
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale35"
type: "Scale"
bottom: "conv35"
top: "conv35"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool5"
type: "Pooling"
bottom: "conv35"
top: "ave_pool5"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route9"
type: "Concat"
bottom: "ave_pool5"
top: "route9"
}
layer {
name: "fc9"
type: "InnerProduct"
bottom: "route9"
top: "fc9"
inner_product_param {
num_output: 168
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu16"
type: "ReLU"
bottom: "fc9"
top: "relu16"
}
layer {
name: "fc10"
type: "InnerProduct"
bottom: "relu16"
top: "fc10"
inner_product_param {
num_output: 672
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power33"
type: "Power"
bottom: "fc10"
top: "power33"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus17"
type: "ReLU6"
bottom: "power33"
top: "relus17"
}
layer {
name: "power34"
type: "Power"
bottom: "relus17"
top: "power34"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route10"
type: "Concat"
bottom: "power34"
top: "route10"
}
layer {
name: "scale5"
type: "Scale"
bottom: "conv35"
bottom: "route10"
top: "scale5"
scale_param {
axis: 0
}
}
layer {
name: "power35"
type: "Power"
bottom: "scale5"
top: "power35"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus18"
type: "ReLU6"
bottom: "power35"
top: "relus18"
}
layer {
name: "power36"
type: "Power"
bottom: "relus18"
top: "power36"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul13"
type: "Eltwise"
bottom: "scale5"
bottom: "power36"
top: "mul13"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv36"
type: "Convolution"
bottom: "mul13"
top: "conv36"
convolution_param {
num_output: 112
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale36"
type: "Scale"
bottom: "conv36"
top: "conv36"
scale_param {
bias_term: true
}
}
layer {
name: "add8"
type: "Eltwise"
bottom: "conv33"
bottom: "conv36"
top: "add8"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv37"
type: "Convolution"
bottom: "add8"
top: "conv37"
convolution_param {
num_output: 672
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale37"
type: "Scale"
bottom: "conv37"
top: "conv37"
scale_param {
bias_term: true
}
}
layer {
name: "power37"
type: "Power"
bottom: "conv37"
top: "power37"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus19"
type: "ReLU6"
bottom: "power37"
top: "relus19"
}
layer {
name: "power38"
type: "Power"
bottom: "relus19"
top: "power38"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul14"
type: "Eltwise"
bottom: "conv37"
bottom: "power38"
top: "mul14"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv38"
type: "DepthwiseConvolution"
bottom: "mul14"
top: "conv38"
convolution_param {
num_output: 672
bias_term: false
pad: 2
kernel_size: 5
group: 672
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale38"
type: "Scale"
bottom: "conv38"
top: "conv38"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool6"
type: "Pooling"
bottom: "conv38"
top: "ave_pool6"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route11"
type: "Concat"
bottom: "ave_pool6"
top: "route11"
}
layer {
name: "fc11"
type: "InnerProduct"
bottom: "route11"
top: "fc11"
inner_product_param {
num_output: 168
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu17"
type: "ReLU"
bottom: "fc11"
top: "relu17"
}
layer {
name: "fc12"
type: "InnerProduct"
bottom: "relu17"
top: "fc12"
inner_product_param {
num_output: 672
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power39"
type: "Power"
bottom: "fc12"
top: "power39"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus20"
type: "ReLU6"
bottom: "power39"
top: "relus20"
}
layer {
name: "power40"
type: "Power"
bottom: "relus20"
top: "power40"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route12"
type: "Concat"
bottom: "power40"
top: "route12"
}
layer {
name: "scale6"
type: "Scale"
bottom: "conv38"
bottom: "route12"
top: "scale6"
scale_param {
axis: 0
}
}
layer {
name: "power41"
type: "Power"
bottom: "scale6"
top: "power41"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus21"
type: "ReLU6"
bottom: "power41"
top: "relus21"
}
layer {
name: "power42"
type: "Power"
bottom: "relus21"
top: "power42"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul15"
type: "Eltwise"
bottom: "scale6"
bottom: "power42"
top: "mul15"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv39"
type: "Convolution"
bottom: "mul15"
top: "conv39"
convolution_param {
num_output: 160
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale39"
type: "Scale"
bottom: "conv39"
top: "conv39"
scale_param {
bias_term: true
}
}
layer {
name: "conv40"
type: "Convolution"
bottom: "conv39"
top: "conv40"
convolution_param {
num_output: 672
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale40"
type: "Scale"
bottom: "conv40"
top: "conv40"
scale_param {
bias_term: true
}
}
layer {
name: "power43"
type: "Power"
bottom: "conv40"
top: "power43"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus22"
type: "ReLU6"
bottom: "power43"
top: "relus22"
}
layer {
name: "power44"
type: "Power"
bottom: "relus22"
top: "power44"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul16"
type: "Eltwise"
bottom: "conv40"
bottom: "power44"
top: "mul16"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv41"
type: "DepthwiseConvolution"
bottom: "mul16"
top: "conv41"
convolution_param {
num_output: 672
bias_term: false
pad: 2
kernel_size: 5
group: 672
stride: 2
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale41"
type: "Scale"
bottom: "conv41"
top: "conv41"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool7"
type: "Pooling"
bottom: "conv41"
top: "ave_pool7"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route13"
type: "Concat"
bottom: "ave_pool7"
top: "route13"
}
layer {
name: "fc13"
type: "InnerProduct"
bottom: "route13"
top: "fc13"
inner_product_param {
num_output: 168
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu18"
type: "ReLU"
bottom: "fc13"
top: "relu18"
}
layer {
name: "fc14"
type: "InnerProduct"
bottom: "relu18"
top: "fc14"
inner_product_param {
num_output: 672
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power45"
type: "Power"
bottom: "fc14"
top: "power45"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus23"
type: "ReLU6"
bottom: "power45"
top: "relus23"
}
layer {
name: "power46"
type: "Power"
bottom: "relus23"
top: "power46"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route14"
type: "Concat"
bottom: "power46"
top: "route14"
}
layer {
name: "scale7"
type: "Scale"
bottom: "conv41"
bottom: "route14"
top: "scale7"
scale_param {
axis: 0
}
}
layer {
name: "power47"
type: "Power"
bottom: "scale7"
top: "power47"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus24"
type: "ReLU6"
bottom: "power47"
top: "relus24"
}
layer {
name: "power48"
type: "Power"
bottom: "relus24"
top: "power48"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul17"
type: "Eltwise"
bottom: "scale7"
bottom: "power48"
top: "mul17"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv42"
type: "Convolution"
bottom: "mul17"
top: "conv42"
convolution_param {
num_output: 160
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale42"
type: "Scale"
bottom: "conv42"
top: "conv42"
scale_param {
bias_term: true
}
}
layer {
name: "conv43"
type: "Convolution"
bottom: "conv42"
top: "conv43"
convolution_param {
num_output: 960
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale43"
type: "Scale"
bottom: "conv43"
top: "conv43"
scale_param {
bias_term: true
}
}
layer {
name: "power49"
type: "Power"
bottom: "conv43"
top: "power49"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus25"
type: "ReLU6"
bottom: "power49"
top: "relus25"
}
layer {
name: "power50"
type: "Power"
bottom: "relus25"
top: "power50"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul18"
type: "Eltwise"
bottom: "conv43"
bottom: "power50"
top: "mul18"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv44"
type: "DepthwiseConvolution"
bottom: "mul18"
top: "conv44"
convolution_param {
num_output: 960
bias_term: false
pad: 2
kernel_size: 5
group: 960
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale44"
type: "Scale"
bottom: "conv44"
top: "conv44"
scale_param {
bias_term: true
}
}
layer {
name: "ave_pool8"
type: "Pooling"
bottom: "conv44"
top: "ave_pool8"
pooling_param {
pool: AVE
global_pooling: true
}
}
layer {
name: "route15"
type: "Concat"
bottom: "ave_pool8"
top: "route15"
}
layer {
name: "fc15"
type: "InnerProduct"
bottom: "route15"
top: "fc15"
inner_product_param {
num_output: 240
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "relu19"
type: "ReLU"
bottom: "fc15"
top: "relu19"
}
layer {
name: "fc16"
type: "InnerProduct"
bottom: "relu19"
top: "fc16"
inner_product_param {
num_output: 960
bias_term: true
weight_filler {
type: "xavier"
}
bias_filler {
type: "constant"
}
}
}
layer {
name: "power51"
type: "Power"
bottom: "fc16"
top: "power51"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus26"
type: "ReLU6"
bottom: "power51"
top: "relus26"
}
layer {
name: "power52"
type: "Power"
bottom: "relus26"
top: "power52"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "route16"
type: "Concat"
bottom: "power52"
top: "route16"
}
layer {
name: "scale8"
type: "Scale"
bottom: "conv44"
bottom: "route16"
top: "scale8"
scale_param {
axis: 0
}
}
layer {
name: "power53"
type: "Power"
bottom: "scale8"
top: "power53"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus27"
type: "ReLU6"
bottom: "power53"
top: "relus27"
}
layer {
name: "power54"
type: "Power"
bottom: "relus27"
top: "power54"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul19"
type: "Eltwise"
bottom: "scale8"
bottom: "power54"
top: "mul19"
eltwise_param {
operation: PROD
}
}
layer {
name: "conv45"
type: "Convolution"
bottom: "mul19"
top: "conv45"
convolution_param {
num_output: 160
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale45"
type: "Scale"
bottom: "conv45"
top: "conv45"
scale_param {
bias_term: true
}
}
layer {
name: "add9"
type: "Eltwise"
bottom: "conv42"
bottom: "conv45"
top: "add9"
eltwise_param {
operation: SUM
}
}
layer {
name: "conv46"
type: "Convolution"
bottom: "add9"
top: "conv46"
convolution_param {
num_output: 960
bias_term: false
pad: 0
kernel_size: 1
group: 1
stride: 1
weight_filler {
type: "xavier"
}
dilation: 1
}
}
layer {
name: "bn_scale46"
type: "Scale"
bottom: "conv46"
top: "conv46"
scale_param {
bias_term: true
}
}
layer {
name: "power55"
type: "Power"
bottom: "conv46"
top: "power55"
power_param {
power: 1.0
scale: 1.0
shift: 3.0
}
}
layer {
name: "relus28"
type: "ReLU6"
bottom: "power55"
top: "relus28"
}
layer {
name: "power56"
type: "Power"
bottom: "relus28"
top: "power56"
power_param {
power: 1.0
scale: 0.1666666716337204
shift: 0.0
}
}
layer {
name: "mul20"
type: "Eltwise"
bottom: "conv46"
bottom: "power56"
top: "mul20"
eltwise_param {
operation: PROD
}
}
layer {
name: "yolo/conv1/dw"
type: "DepthwiseConvolution"
bottom: "mul20"
top: "yolo/conv1/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 960
bias_term: false
pad: 1
kernel_size: 3
group: 960
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv1/dw/scale"
type: "Scale"
bottom: "yolo/conv1/dw"
top: "yolo/conv1/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv1/dw/relu"
type: "ReLU"
bottom: "yolo/conv1/dw"
top: "yolo/conv1/dw"
}
layer {
name: "yolo/conv1"
type: "Convolution"
bottom: "yolo/conv1/dw"
top: "yolo/conv1"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 672
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv1/scale"
type: "Scale"
bottom: "yolo/conv1"
top: "yolo/conv1"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv1/relu"
type: "ReLU"
bottom: "yolo/conv1"
top: "yolo/conv1"
}
layer {
name: "upsample"
type: "Deconvolution"
bottom: "yolo/conv1"
top: "upsample"
param { lr_mult: 0 decay_mult: 0 }
convolution_param {
num_output: 672
group: 672
kernel_size: 1 stride: 2 pad: 0
weight_filler: {
type: "constant"
value : 1
}
bias_term: false
}
}
layer {
top: "maxpool"
name: "maxpool"
bottom: "upsample"
type: "Pooling"
pooling_param {
kernel_size: 2
stride: 1
pool: MAX
pad: 1
}
}
layer {
name: "yolo/conv2/dw"
type: "DepthwiseConvolution"
bottom: "mul16"
top: "yolo/conv2/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 672
bias_term: false
pad: 1
kernel_size: 3
group: 672
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv2/dw/scale"
type: "Scale"
bottom: "yolo/conv2/dw"
top: "yolo/conv2/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv2/dw/relu"
type: "ReLU"
bottom: "yolo/conv2/dw"
top: "yolo/conv2/dw"
}
layer {
name: "yolo/conv2"
type: "Convolution"
bottom: "yolo/conv2/dw"
top: "yolo/conv2"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 672
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv2/scale"
type: "Scale"
bottom: "yolo/conv2"
top: "yolo/conv2"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv2/relu"
type: "ReLU"
bottom: "yolo/conv2"
top: "yolo/conv2"
}
layer {
name: "yolo/conv2/sum"
type: "Eltwise"
bottom: "maxpool"
bottom: "yolo/conv2"
top: "yolo/conv2/sum"
}
layer {
name: "yolo/conv3/dw"
type: "DepthwiseConvolution"
bottom: "yolo/conv2/sum"
top: "yolo/conv3/dw"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 672
bias_term: false
pad: 1
kernel_size: 3
group: 672
engine: CAFFE
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv3/dw/scale"
type: "Scale"
bottom: "yolo/conv3/dw"
top: "yolo/conv3/dw"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv3/dw/relu"
type: "ReLU"
bottom: "yolo/conv3/dw"
top: "yolo/conv3/dw"
}
layer {
name: "yolo/conv3"
type: "Convolution"
bottom: "yolo/conv3/dw"
top: "yolo/conv3"
param {
lr_mult: 1
decay_mult: 1
}
convolution_param {
num_output: 672
bias_term: false
kernel_size: 1
weight_filler {
type: "msra"
}
}
}
layer {
name: "yolo/conv3/scale"
type: "Scale"
bottom: "yolo/conv3"
top: "yolo/conv3"
param {
lr_mult: 1
decay_mult: 0.0
}
param {
lr_mult: 2
decay_mult: 0.0
}
scale_param {
filler {
value: 1
}
bias_term: true
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv3/relu"
type: "ReLU"
bottom: "yolo/conv3"
top: "yolo/conv3"
}
layer {
name: "yolo/conv4"
type: "Convolution"
bottom: "yolo/conv1"
top: "yolo/conv4"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 18
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "yolo/conv5"
type: "Convolution"
bottom: "yolo/conv3"
top: "yolo/conv5"
param {
lr_mult: 1
decay_mult: 1
}
param {
lr_mult: 2
decay_mult: 0
}
convolution_param {
num_output: 18
kernel_size: 1
pad: 0
stride: 1
weight_filler {
type: "msra"
}
bias_filler {
value: 0
}
}
}
layer {
name: "detection_out"
type: "Yolov3DetectionOutput"
bottom: "yolo/conv4"
bottom: "yolo/conv5"
top: "detection_out"
yolov3_detection_output_param {
confidence_threshold: 0.01
nms_threshold: 0.45
num_classes: 1
#10,14, 23,27, 37,58, 81,82, 135,169, 344,319
biases: 18
biases: 34
biases: 45
biases: 84
biases: 65
biases: 181
biases: 141
biases: 111
biases: 129
biases: 241
biases: 254
biases: 254
mask:3
mask:4
mask:5
mask:0
mask:1
mask:2
anchors_scale:32
anchors_scale:16
mask_group_num:2
}
}
Non merge_bn model
first, thank you for your contribute
can you provide a Non merge_bn MobileNet-YOLOv3-Lite model ?
How do i transform image data to lmdb on windows
hello:
I want to train my own yolov3 model, I do't know how to get the lmdb data.
In caffe, it has converts_imageset, but I doubt it may not be suitable for yolo3 or ssd train data.
So How do i transform image data to lmdb on windows
hi,thanks for your work, but mAp is too bad, how can i improve the result?
anchor out of bound
First, thanks for your sharing.
Second, It seems that there is an anchor out of bound issue at line 514 in region_loss_layer.cpp for one biases in prototxt is 16.62, which is larger than "side_".
hi,thanks for your great job, but my detected calss does't agree with yours
支持Win32吗？
Build wrong in gpu,but fine in cpu only
hi!
build_win.cmd in win10 with GPU evn, but wrong below：
C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\V120\Microsoft.CppCommon.targets(170,5): error MSB6006: “cmd.exe
if build in cpu only, it is right!
where is wrong?
Thank you
楼主，你这个问题最终怎么解决的呢？
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.