cuda8+cuDNN Faster R-CNN安装塈运行demo

迷南。 2022-09-29 04:26 142阅读 0赞

# 安装cuda #

cuda8安装参见网上教程

# 安装cuDNN #

py-faster-rcnn/caffe-fast-rcnn目前不支持cuDNN5。  
如果使用cuDNN5，在编译caffe-fast-rcnn时会报错：

> error: argument of type “int” is incompatible with parameter of type “cudnnNanPropagation\_t”  
> 解决编译错误的方法参见我的另一篇博客：  
> [《使用cuDNN5编译py-faster-rcnn错误：cudnn.hpp(126): error: argument of type “int” is incompatible …》][cuDNN5_py-faster-rcnn_cudnn.hpp_126_ error_ argument of type _int_ is incompatible]

如果使用cuDNN4就不会有编译错误。  
cuDNN安装方式(cuDNN4为例)

tar zxvf cudnn-7.0-linux-x64-v4.0-prod.tgz 
    sudo cp cuda/include/cudnn.h /usr/local/cuda/include 
    sudo cp -d cuda/lib64/libcudnn* /usr/local/cuda/lib64

# cuDNN4 OR 5？ #

那么应该安装cuDNN4还是cuDNN5呢？  
这取决于你的NVIDIA显卡与cuDNN之间是否有兼容性问题。  
如果你的电脑上用cuDNN4能正常执行demo，那自然好，  
如果不能，就换cuDNN5试试，原因参见我的另一篇博客  
[《cuDNN兼容性问题造成的caffe/mnist,py-faster-rcnn/demo运行结果错误》][cuDNN_caffe_mnist_py-faster-rcnn_demo]

# 下载faster r-cnn代码 #

$git clone --recursive https://github.com/rbgirshick/py-faster-rcnn.git

# 下载模型数据 #

模型数据有点大，要花点时间

$./data/scripts/fetch_faster_rcnn_models.sh

# 编译caffe #

这里的caffe是Faster R-CNN自带的caffe，在caffe-fast-rcnn文件夹下

$cd caffe-fast-rcnn
    #如果没有cmake，要安装cmake
    #$sudo apt-get install cmake
    # 执行cmake生成Makefile
    #编译CPU版本
    #cmake -DCPU_ONLY=ON -DBLAS=Open .
    #编译GPU版本
    $cmake -DBLAS=Open -DCUDA_NVCC_FLAGS=--Wno-deprecated-gpu-targets .
    $make -j 8

> 关于编译caffe的细节参见我的另一篇博客 [《Ubuntu16:cmake生成Makefile编译caffe过程(OpenBLAS/CPU+GPU)塈解决nvcc warning:The ‘compute\_20’, ‘sm\_20’》][Ubuntu16_cmake_Makefile_caffe_OpenBLAS_CPU_GPU_nvcc warning_The _compute_20_ _sm_20]

# 安装pycaffe的依赖库 #

caffe-fast-rcnn/python/requirement.txt文件列出了pycaffe的所有依赖库，要把requirements下的依赖都装一遍

> guyadong@10km-ubuntu:~/tmp/py-faster-rcnn/caffe-fast-rcnn/python$ cat requirements.txt  
> Cython>=0.19.2  
> numpy>=1.7.1  
> scipy>=0.13.2  
> scikit-image>=0.9.3  
> matplotlib>=1.3.1  
> ipython>=3.0.0  
> h5py>=2.2.0  
> leveldb>=0.191  
> networkx>=1.8.1  
> nose>=1.3.0  
> pandas>=0.12.0  
> python-dateutil>=1.4,<2  
> protobuf>=2.5.0  
> python-gflags>=2.0  
> pyyaml>=3.10  
> Pillow>=2.3.0

pycaffe的依赖库是用Python的包管理工具pip来安装的,如果还没安装pip，则要先安装python-pip

$ sudo apt-get install python-pip
    $ sudo apt-get install python-opencv
    $ sudp apt-get install python-tk
    $ sudo apt-get install python-dateutil
    $ cd caffe-fast-rcnn/python
    $ for req in $(cat requirements.txt); do sudo pip install $req; done
    $ sudo pip install easydict

# 编译faster-rcnn #

编译好caffe-fast-rcnn后，在py-faster-rcnn/lib中执行make命令：

> $ make  
> running build  
> running build\_ext  
> skipping ‘utils/bbox.c’ Cython extension (up-to-date)  
> building ‘utils.cython\_bbox’ extension  
> creating build  
> creating build/temp.linux-x86\_64-2.7  
> creating build/temp.linux-x86\_64-2.7/utils  
> x86\_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -I/usr/lib/python2.7/dist-packages/numpy/core/include -I/usr/include/python2.7 -c utils/bbox.c -o build/temp.linux-x86\_64-2.7/utils/bbox.o -Wno-cpp -Wno-unused-function  
> creating build/lib.linux-x86\_64-2.7  
> creating build/lib.linux-x86\_64-2.7/utils  
> x86\_64-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security build/temp.linux-x86\_64-2.7/utils/bbox.o -o build/lib.linux-x86\_64-2.7/utils/cython\_bbox.so  
> skipping ‘nms/cpu\_nms.c’ Cython extension (up-to-date)  
> building ‘nms.cpu\_nms’ extension  
> creating build/temp.linux-x86\_64-2.7/nms  
> x86\_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -I/usr/lib/python2.7/dist-packages/numpy/core/include -I/usr/include/python2.7 -c nms/cpu\_nms.c -o build/temp.linux-x86\_64-2.7/nms/cpu\_nms.o -Wno-cpp -Wno-unused-function  
> creating build/lib.linux-x86\_64-2.7/nms  
> x86\_64-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security build/temp.linux-x86\_64-2.7/nms/cpu\_nms.o -o build/lib.linux-x86\_64-2.7/nms/cpu\_nms.so  
> skipping ‘nms/gpu\_nms.cpp’ Cython extension (up-to-date)  
> building ‘nms.gpu\_nms’ extension  
> /usr/local/cuda-8.0/bin/nvcc -I/usr/lib/python2.7/dist-packages/numpy/core/include -I/usr/local/cuda-8.0/include -I/usr/include/python2.7 -c nms/nms\_kernel.cu -o build/temp.linux-x86\_64-2.7/nms/nms\_kernel.o -arch=sm\_35 –ptxas-options=-v -c –compiler-options ‘-fPIC’  
> ptxas info : 0 bytes gmem  
> ptxas info : Compiling entry function ‘\_Z10nms\_kernelifPKfPy’ for ‘sm\_35’  
> ptxas info : Function properties for \_Z10nms\_kernelifPKfPy  
> 0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads  
> ptxas info : Used 25 registers, 1280 bytes smem, 344 bytes cmem\[0\], 8 bytes cmem[2][]  
> x86\_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -I/usr/lib/python2.7/dist-packages/numpy/core/include -I/usr/local/cuda-8.0/include -I/usr/include/python2.7 -c nms/gpu\_nms.cpp -o build/temp.linux-x86\_64-2.7/nms/gpu\_nms.o -Wno-unused-function  
> cc1plus: warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++  
> In file included from /usr/lib/python2.7/dist-packages/numpy/core/include/numpy/ndarraytypes.h:1777:0,  
> from /usr/lib/python2.7/dist-packages/numpy/core/include/numpy/ndarrayobject.h:18,  
> from /usr/lib/python2.7/dist-packages/numpy/core/include/numpy/arrayobject.h:4,  
> from nms/gpu\_nms.cpp:449:  
> /usr/lib/python2.7/dist-packages/numpy/core/include/numpy/npy\_1\_7\_deprecated\_api.h:15:2: warning: \#warning “Using deprecated NumPy API, disable it by ” “\#defining NPY\_NO\_DEPRECATED\_API NPY\_1\_7\_API\_VERSION” \[-Wcpp\]  
> \#warning “Using deprecated NumPy API, disable it by ” \\  
> ^  
> c++ -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security build/temp.linux-x86\_64-2.7/nms/nms\_kernel.o build/temp.linux-x86\_64-2.7/nms/gpu\_nms.o -L/usr/local/cuda-8.0/lib64 -Wl,-R/usr/local/cuda-8.0/lib64 -lcudart -o build/lib.linux-x86\_64-2.7/nms/gpu\_nms.so  
> skipping ‘pycocotools/\_mask.c’ Cython extension (up-to-date)  
> building ‘pycocotools.\_mask’ extension  
> creating build/temp.linux-x86\_64-2.7/pycocotools  
> x86\_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -I/usr/lib/python2.7/dist-packages/numpy/core/include -Ipycocotools -I/usr/include/python2.7 -c pycocotools/maskApi.c -o build/temp.linux-x86\_64-2.7/pycocotools/maskApi.o -Wno-cpp -Wno-unused-function -std=c99  
> x86\_64-linux-gnu-gcc -pthread -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -fno-strict-aliasing -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -fPIC -I/usr/lib/python2.7/dist-packages/numpy/core/include -Ipycocotools -I/usr/include/python2.7 -c pycocotools/\_mask.c -o build/temp.linux-x86\_64-2.7/pycocotools/\_mask.o -Wno-cpp -Wno-unused-function -std=c99  
> creating build/lib.linux-x86\_64-2.7/pycocotools  
> x86\_64-linux-gnu-gcc -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -fno-strict-aliasing -DNDEBUG -g -fwrapv -O2 -Wall -Wstrict-prototypes -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security -Wl,-Bsymbolic-functions -Wl,-z,relro -Wdate-time -D\_FORTIFY\_SOURCE=2 -g -fstack-protector-strong -Wformat -Werror=format-security build/temp.linux-x86\_64-2.7/pycocotools/maskApi.o build/temp.linux-x86\_64-2.7/pycocotools/\_mask.o -o build/lib.linux-x86\_64-2.7/pycocotools/\_mask.so

# 运行demo #

$python tools/demo.py

> ….  
> I0314 16:11:46.644143 22175 net.cpp:270\] This network produces output bbox\_pred  
> I0314 16:11:46.644161 22175 net.cpp:270\] This network produces output cls\_prob  
> I0314 16:11:46.644192 22175 net.cpp:283\] Network initialization done.  
> \[libprotobuf WARNING google/protobuf/io/coded\_stream.cc:537\] Reading dangerously large protocol message. >If the message turns out to be larger than 2147483647 bytes, parsing will be halted for security reasons. To >increase the limit (or to disable these warnings), see CodedInputStream::SetTotalBytesLimit() in >google/protobuf/io/coded\_stream.h.  
> \[libprotobuf WARNING google/protobuf/io/coded\_stream.cc:78\] The total number of bytes read was 548317115  
> I0314 16:11:46.803853 22175 net.cpp:816\] Ignoring source layer data  
> I0314 16:11:46.873862 22175 net.cpp:816\] Ignoring source layer drop6  
> I0314 16:11:46.883658 22175 net.cpp:816\] Ignoring source layer drop7  
> I0314 16:11:46.883672 22175 net.cpp:816\] Ignoring source layer fc7\_drop7\_0\_split  
> I0314 16:11:46.883926 22175 net.cpp:816\] Ignoring source layer loss\_cls  
> I0314 16:11:46.883936 22175 net.cpp:816\] Ignoring source layer loss\_bbox  
> I0314 16:11:46.885593 22175 net.cpp:816\] Ignoring source layer silence\_rpn\_cls\_score  
> I0314 16:11:46.885612 22175 net.cpp:816\] Ignoring source layer silence\_rpn\_bbox\_pred  
> Loaded network /home/guyadong/tmp/py-faster-rcnn/data/faster\_rcnn\_models/VGG16\_faster\_rcnn\_final.caffemodel  
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> Demo for data/demo/000456.jpg  
> Detection took 2.097s for 300 object proposals  
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> Demo for data/demo/000542.jpg  
> Detection took 1.695s for 161 object proposals  
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> Demo for data/demo/001150.jpg  
> Detection took 1.908s for 194 object proposals  
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> Demo for data/demo/001763.jpg  
> Detection took 1.785s for 196 object proposals  
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  
> Demo for data/demo/004545.jpg  
> Detection took 1.881s for 300 object proposals

如果在CPU上运行，要复杂一些，需要手工修改一些脚本，参见[《Faster R-CNN CPU环境搭建》][2].

# 参考资料 #

[《Faster R-CNN CPU环境搭建》][2]  
[《Faster R-CNN安装笔记，只用CPU 》][Faster R-CNN_CPU]

[cuDNN5_py-faster-rcnn_cudnn.hpp_126_ error_ argument of type _int_ is incompatible]: http://blog.csdn.net/10km/article/details/62418583
[cuDNN_caffe_mnist_py-faster-rcnn_demo]: http://blog.csdn.net/10km/article/details/62421445
[Ubuntu16_cmake_Makefile_caffe_OpenBLAS_CPU_GPU_nvcc warning_The _compute_20_ _sm_20]: http://blog.csdn.net/10km/article/details/61619573
[2]: http://www.cnblogs.com/justinzhang/p/5386837.html
[Faster R-CNN_CPU]: http://blog.sina.com.cn/s/blog_679f93560102wpyf.html