公司做网站 分录,揭阳网站建设策划方案,个人博客网站模板,营销网站建站以下仅仅为一个总结#xff0c;参考了网上的众多资料#xff0c;仅备忘记。
主要链接
deeplab主页#xff1a;http://liangchiehchen.com/projects/DeepLab.html官方代码#xff1a;https://bitbucket.org/aquariusjay/deeplab-public-ver2python 版caffe实现#xff1a…以下仅仅为一个总结参考了网上的众多资料仅备忘记。
主要链接
deeplab主页http://liangchiehchen.com/projects/DeepLab.html官方代码https://bitbucket.org/aquariusjay/deeplab-public-ver2python 版caffe实现https://github.com/TheLegendAli/DeepLab-Context2 model下载http://liangchiehchen.com/projects/DeepLab_Models.htmlDeepLabv2_VGG16 预训练模型DeepLabv2_ResNet101预训练模型pytorch实现的deeplabhttps://github.com/isht7/pytorch-deeplab-resnet 网上开源代码martinkersner/train-DeepLab
主要运行步骤
如下我们主要以网上开源的一个版本讲解https://github.com/xmojiao/deeplab_v2 . 主要的步骤可以参考 1. 图像语义分割从头开始训练deeplab v2系列之一【源码解析】 2. 图像语义分割从头开始训练deeplab v2系列之二【VOC2012数据集】 3. Deeplab v2 调试全过程Ubuntu 16.04cuda8.0
下面说一些这上面没有的我遇到的一些问题。 1. 安装 matio: 上面资料中都使用 matio-1.5.2.tar.gz 但是我装不上可能与我的库不兼容于是我下载了最新的 matio-1.5.11 并且按照如下命令运行安装
cd matio-1.5.11
./configure --prefix/data1/... (填写自己的安装目录)
make
make check 可略
make install 最后在bash.rc上加上 LD_LIBRARY_PATH/your/path/to/libmatio.so.2
参考http://blog.csdn.net/houqiqi/article/details/46469981 2. 使用的Caffe版本比较陈旧导致会出现很多和最新环境不兼容的情况我使用的是cuDNN6.0,cuda8.0 出现
./include/caffe/util/cudnn.hpp: In function ‘void caffe::cudnn::createPoolingDesc(cudnnPoolingStruct**, caffe::PoolingParameter_PoolMethod, cudnnPoolingMode_t*, int, int, int, int, int, int)’:
./include/caffe/util/cudnn.hpp:127:41: error: too few arguments to function ‘cudnnStatus_t cudnnSetPooling2dDescriptor(cudnnPoolingDescriptor_t, cudnnPoolingMode_t, cudnnNanPropagation_t, int, int, int, int, int, int)’pad_h, pad_w, stride_h, stride_w));
这是由于所使用的cuDNN版本不一致的导致的作者配置环境是cuDNN 4.0但是5.0版本后的cuDNN接口有所变化。
解决方法 将以下几个文件用最新BVLC版本的caffe对应文件替换并重新编译
./include/caffe/util/cudnn.hpp
./include/caffe/layers/cudnn_conv_layer.hpp
./include/caffe/layers/cudnn_relu_layer.hpp
./include/caffe/layers/cudnn_sigmoid_layer.hpp
./include/caffe/layers/cudnn_tanh_layer.hpp./src/caffe/layers/cudnn_conv_layer.cpp
./src/caffe/layers/cudnn_conv_layer.cu
./src/caffe/layers/cudnn_relu_layer.cpp
./src/caffe/layers/cudnn_relu_layer.cu
./src/caffe/layers/cudnn_sigmoid_layer.cpp
./src/caffe/layers/cudnn_sigmoid_layer.cu
./src/caffe/layers/cudnn_tanh_layer.cpp
./src/caffe/layers/cudnn_tanh_layer.cu
参考http://blog.csdn.net/tianrolin/article/details/71246472 3. 如何解决deeplab v2识别结果为全黑图像的问题 正如作者而言http://liangchiehchen.com/projects/DeepLab_FAQ.html
Q: When evaluating the DeepLab outputs (without CRF), I got all-background results (i.e., all black results). Is there anything wrong?A: Please double check if the name of your fc8 is fc8_voc12 in the generated test_val.prototxt or test_test.prototxt (after running run_pascal.sh). The name should be matched for initialization.
主要是作者预训练的模型与你实际测试的模型出现了偏差。主要是fc8 的问题。 以https://github.com/xmojiao/deeplab_v2 为例如果直接使用此代码则run_pascal.sh出现在voc12目录下而run_pascal.sh上EXP2. 这与官方的预设不同。官方代码假定run_pascal.sh应该出现在voc12上一级目录下。 这样最后测试时会出现fc8_voc12_1,fc8_voc12_2,fc8_voc12_3,fc8_voc12_4被忽略的情况。
另外由于测试时需要将test.prototxt复制为test_val.prototxt所以应该主要修改test.prototxt。 test.prototxt改动如下
layer {name: datatype: ImageSegDatatop: datatop: labeltop: data_diminclude {phase: TEST}transform_param {mirror: falsecrop_size: 513mean_value: 104.008mean_value: 116.669mean_value: 122.675}image_data_param {root_folder: ${DATA_ROOT}source: ../${EXP}/list/${TEST_SET}.txt 改变batch_size: 1label_type: NONE}
}
run_pascal.sh改变如下
#!/bin/sh## MODIFY PATH for YOUR SETTING
ROOT_DIR/data1/caiyong.wang/data/deeplab_dataCAFFE_DIR../deeplab-public-ver2
CAFFE_BIN${CAFFE_DIR}/.build_release/tools/caffe.binEXPvoc12 #适应原始训练好的模型目录 改变
EXP2. #当前目录下 改变if [ ${EXP2} . ]; thenNUM_LABELS21DATA_ROOT${ROOT_DIR}/VOC_aug/dataset/
elseNUM_LABELS0echo Wrong EXP name
fi## Specify which model to train
########### voc12 ################
NET_IDdeeplab_largeFOV## Variables used for weakly or semi-supervisedly training
#TRAIN_SET_SUFFIX
TRAIN_SET_SUFFIX_aug#TRAIN_SET_STRONGtrain
#TRAIN_SET_STRONGtrain200
#TRAIN_SET_STRONGtrain500
#TRAIN_SET_STRONGtrain1000
#TRAIN_SET_STRONGtrain750#TRAIN_SET_WEAK_LEN5000DEV_ID0####### Create dirsCONFIG_DIR${EXP2}/config/${NET_ID}
MODEL_DIR${EXP2}/model/${NET_ID}
mkdir -p ${MODEL_DIR}
LOG_DIR${EXP2}/log/${NET_ID}
mkdir -p ${LOG_DIR}
export GLOG_log_dir${LOG_DIR}## RunRUN_TRAIN0
RUN_TEST1
RUN_TRAIN20
RUN_TEST20## Training #1 (on train_aug)if [ ${RUN_TRAIN} -eq 1 ]; then#LIST_DIR${EXP2}/listTRAIN_SETtrain${TRAIN_SET_SUFFIX}if [ -z ${TRAIN_SET_WEAK_LEN} ]; thenTRAIN_SET_WEAK${TRAIN_SET}_diff_${TRAIN_SET_STRONG}comm -3 ${LIST_DIR}/${TRAIN_SET}.txt ${LIST_DIR}/${TRAIN_SET_STRONG}.txt ${LIST_DIR}/${TRAIN_SET_WEAK}.txtelseTRAIN_SET_WEAK${TRAIN_SET}_diff_${TRAIN_SET_STRONG}_head${TRAIN_SET_WEAK_LEN}comm -3 ${LIST_DIR}/${TRAIN_SET}.txt ${LIST_DIR}/${TRAIN_SET_STRONG}.txt | head -n ${TRAIN_SET_WEAK_LEN} ${LIST_DIR}/${TRAIN_SET_WEAK}.txtfi#MODEL${EXP2}/model/${NET_ID}/init.caffemodel#echo Training net ${EXP2}/${NET_ID}for pname in train solver; dosed $(eval echo $(cat sub.sed)) \${CONFIG_DIR}/${pname}.prototxt ${CONFIG_DIR}/${pname}_${TRAIN_SET}.prototxtdoneCMD${CAFFE_BIN} train \--solver${CONFIG_DIR}/solver_${TRAIN_SET}.prototxt \--gpu${DEV_ID}if [ -f ${MODEL} ]; thenCMD${CMD} --weights${MODEL}fiecho Running ${CMD} ${CMD}
fi## Test #1 specification (on val or test)if [ ${RUN_TEST} -eq 1 ]; then#for TEST_SET in val; doTEST_ITERcat ${EXP2}/list/${TEST_SET}.txt | wc -lMODEL${EXP2}/model/${NET_ID}/test.caffemodelif [ ! -f ${MODEL} ]; thenMODELls -t ${EXP2}/model/${NET_ID}/train_iter_*.caffemodel | head -n 1fi#echo Testing net ${EXP2}/${NET_ID}FEATURE_DIR${EXP2}/features/${NET_ID}mkdir -p ${FEATURE_DIR}/${TEST_SET}/fc8mkdir -p ${FEATURE_DIR}/${TEST_SET}/fc9mkdir -p ${FEATURE_DIR}/${TEST_SET}/seg_scoresed $(eval echo $(cat sub.sed)) \${CONFIG_DIR}/test.prototxt ${CONFIG_DIR}/test_${TEST_SET}.prototxtCMD${CAFFE_BIN} test \--model${CONFIG_DIR}/test_${TEST_SET}.prototxt \--weights${MODEL} \--gpu${DEV_ID} \--iterations${TEST_ITER}echo Running ${CMD} ${CMD}done
fi## Training #2 (finetune on trainval_aug)if [ ${RUN_TRAIN2} -eq 1 ]; then#LIST_DIR${EXP2}/listTRAIN_SETtrainval${TRAIN_SET_SUFFIX}if [ -z ${TRAIN_SET_WEAK_LEN} ]; thenTRAIN_SET_WEAK${TRAIN_SET}_diff_${TRAIN_SET_STRONG}comm -3 ${LIST_DIR}/${TRAIN_SET}.txt ${LIST_DIR}/${TRAIN_SET_STRONG}.txt ${LIST_DIR}/${TRAIN_SET_WEAK}.txtelseTRAIN_SET_WEAK${TRAIN_SET}_diff_${TRAIN_SET_STRONG}_head${TRAIN_SET_WEAK_LEN}comm -3 ${LIST_DIR}/${TRAIN_SET}.txt ${LIST_DIR}/${TRAIN_SET_STRONG}.txt | head -n ${TRAIN_SET_WEAK_LEN} ${LIST_DIR}/${TRAIN_SET_WEAK}.txtfi#MODEL${EXP2}/model/${NET_ID}/init2.caffemodelif [ ! -f ${MODEL} ]; thenMODELls -t ${EXP2}/model/${NET_ID}/train_iter_*.caffemodel | head -n 1fi#echo Training2 net ${EXP2}/${NET_ID}for pname in train solver2; dosed $(eval echo $(cat sub.sed)) \${CONFIG_DIR}/${pname}.prototxt ${CONFIG_DIR}/${pname}_${TRAIN_SET}.prototxtdoneCMD${CAFFE_BIN} train \--solver${CONFIG_DIR}/solver2_${TRAIN_SET}.prototxt \--weights${MODEL} \--gpu${DEV_ID}echo Running ${CMD} ${CMD}
fi## Test #2 on official test setif [ ${RUN_TEST2} -eq 1 ]; then#for TEST_SET in val test; doTEST_ITERcat ${EXP2}/list/${TEST_SET}.txt | wc -lMODEL${EXP2}/model/${NET_ID}/test2.caffemodelif [ ! -f ${MODEL} ]; thenMODELls -t ${EXP2}/model/${NET_ID}/train2_iter_*.caffemodel | head -n 1fi#echo Testing2 net ${EXP2}/${NET_ID}FEATURE_DIR${EXP2}/features2/${NET_ID}mkdir -p ${FEATURE_DIR}/${TEST_SET}/fc8mkdir -p ${FEATURE_DIR}/${TEST_SET}/crfsed $(eval echo $(cat sub.sed)) \${CONFIG_DIR}/test.prototxt ${CONFIG_DIR}/test_${TEST_SET}.prototxtCMD${CAFFE_BIN} test \--model${CONFIG_DIR}/test_${TEST_SET}.prototxt \--weights${MODEL} \--gpu${DEV_ID} \--iterations${TEST_ITER}echo Running ${CMD} ${CMD}done
fi
