1 添加必要的库
cython,python-opencv,easydict
$ sudo apt-get install python-pip
$ sudo pip install cython
$ sudo apt-get install python-opencv
$ sudo pip install easydict
2 下载源代码 py-faster-rcnn
$ cd ~/software/caffe
$ git clone --recursive https://github.com/rbgirshick/py-faster-rcnn.git
$ cp -r py-faster-rcnn/ ~/
注意,--recursive 一定要有,否则 py-faster-rcnn 目录下没有 caffe-fast-rcnn 文件夹。
3 修改~/py-faster-rcnn/caffe-fast-rcnn目录下的Makefile和Makefile.config
直接把之前 caffe 目录下的 Makefile 和 Makefile.config 复制到 caffe-fast-rcnn 下,
$ cp ~/caffe/Makefile ~/py-faster-rcnn/caffe-fast-rcnn/
$ cp ~/caffe/Makefile.config ~/py-faster-rcnn/caffe-fast-rcnn/
4 编译~/py-faster-rcnn下的lib文件
4.1 修改py-faster-rcnn/lib/setup.py文件
第135行
'nvcc': ['-arch=sm_35',
改为：自己显卡的计算能力
'nvcc': ['-arch=sm_20',
(不改好像也可以)
4.2 编译 lib文件
$ cd ~/py-faster-rcnn/lib
$ make
5 编译py-faster-rcnn下的caffe-fast-rcnn
$ cd ~/py-faster-rcnn/caffe-fast-rcnn/
$ make
此时会报错,原因在于 faster rcnn 采用的是 cudnn_v4,而我们安装的是 cudnn_v6,解决方法
如下:
(1) 用最新 caffe 源码的以下文件替换掉 faster rcnn 的对应文件
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_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
$ cp ~/caffe/include/caffe/layers/cudnn_relu_layer.hpp ~/py-faster-rcnn/caffe-fast-rcnn/include/caffe/layers
$ cp ~/caffe/include/caffe/layers/cudnn_sigmoid_layer.hpp ~/py-faster-rcnn/caffe-fast-rcnn/include/caffe/layers
$ cp ~/caffe/include/caffe/layers/cudnn_tanh_layer.hpp ~/py-faster-rcnn/caffe-fast-rcnn/include/caffe/layers
$ cp ~/caffe/src/caffe/layers/cudnn_relu_layer.cpp ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers
$ cp ~/caffe/src/caffe/layers/cudnn_relu_layer.cu ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers
$ cp ~/caffe/src/caffe/layers/cudnn_sigmoid_layer.cpp ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers
$ cp ~/caffe/src/caffe/layers/cudnn_tanh_layer.cpp ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers
$ cp ~/caffe/src/caffe/layers/cudnn_tanh_layer.cu ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers
(2) 用 caffe 源码中的这个文件替换掉 faster rcnn 对应文件
include/caffe/util/cudnn.hpp
$ cp ~/caffe/include/caffe/util/cudnn.hpp ~/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util
(3) 将 faster rcnn 中的 src/caffe/layers/cudnn_conv_layer.cu 文件中的所有
cudnnConvolutionBackwardData_v3 函数名替换为
cudnnConvolutionBackwardData
cudnnConvolutionBackwardFilter_v3 函数名替换为
cudnnConvolutionBackwardFilter
vim ~/py-faster-rcnn/caffe-fast-rcnn/src/caffe/layers/cudnn_conv_layer.cu
运行指令clean，再编译
$ cd ~/py-faster-rcnn/caffe-fast-rcnn/
$ make clean
$ make –j4
$ make pycaffe
出现警告：
nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
暂时忽略。
6 下载faster_rcnn_models.tgz放在~/py-faster-rcnn/data并解压
编译好之后,需要下载一个 faster_rcnn_models.tgz 压缩包
$ cd ~/py-faster-rcnn
$ ./data/scripts/fetch_faster_rcnn_models.sh
注意：最好再windows上下载。用命令下载，容易失败，所有就换到windows下载，再复制到相应目录解压
下载地址：http://www.cs.berkeley.edu/%7Erbg/faster-rcnn-data/faster_rcnn_models.tgz
或者
https://dl.dropboxusercontent.com/s/o6ii098bu51d139/faster_rcnn_models.tgz?dl=0
用https://mm.ww.rrjs.pw/翻墙下载即可
下载好后,将压缩包放在 py-faster-rcnn/data/下解压即可
$ tar -zxvf faster_rcnn_models.tgz -C ~/py-faster-rcnn/data
$ cd ~/py-faster-rcnn/tools/
7 运行~/py-faster-rcnn/tools下的demo.py
$ cd ~/py-faster-rcnn/tools
$ ./demo.py
问题：
$ ./demo.py
Traceback (most recent call last):
File "./demo.py", line 17, in <module>
from fast_rcnn.config import cfg
File "/home/fc/py-faster-rcnn/tools/../lib/fast_rcnn/config.py", line 23, in <module>
from easydict import EasyDict as edict
ImportError: No module named easydict
解决：
$ sudo pip install easydict
其实easydict前面已经安装过了，所有再执行这个命令还是不能解决问题。不断尝试后发现，切换到root用户就没有这个问题了。
但出现新的问题
File "/usr/lib/python2.7/lib-tk/Tkinter.py", line 42, in <module>
raise ImportError, str(msg) + ', please install the python-tk package'
ImportError: No module named _tkinter, please install the python-tk package
这个问题到比较容易，缺python-tk那就安装：
$ sudo apt-get install python-tk
新问题：
Loaded network /home/fc/py-faster-rcnn/data/faster_rcnn_models/VGG16_faster_rcnn_final.caffemodel
F0105 20:51:57.321182 12180 relu_layer.cu:26] Check failed: error == cudaSuccess (9 vs. 0) invalid configuration argument
*** Check failure stack trace: ***
已放弃 (核心已转储)
解决1：
网上（https://www.zhihu.com/question/41493725）说这个原因可能显存不足 ，可以先使用小点的模型比如ZF来进行检测，执行demo.py时加参数--net zf，或者直接修改demo.py里面的代码将默认加载的model改为zf.如果想修改batch_size进行训练，可以在faster-rcnn-root/lib/fast_rcnn/config.py中进行修改。
所以执行：
$ ./demo.py --net zf
还是，不过“9 vs. 0”变成“8 vs. 0”
Loaded network /home/fc/py-faster-rcnn/data/faster_rcnn_models/ZF_faster_rcnn_final.caffemodel
F0105 20:57:31.931417 12305 roi_pooling_layer.cu:91] Check failed: error == cudaSuccess (8 vs. 0) invalid device function
*** Check failure stack trace: ***
已放弃 (核心已转储)
解决2：
还有地方（http://blog.csdn.net/u013078356/article/details/51009470）说，是因为显卡的计算能力不同而导致的，修改 ~/py-faster-rcnn/lib//setup.py 的第135行，将arch改为与你显卡相匹配的数值，（比如我的GTX 760，计算能力是3.0，就将sm_35改成了sm_30，前面我已经改过了）然后删除 py-faster-rcnn/lib/utils/bbox.c，py-faster-rcnn/lib/nms/cpu_nms.c ，py-faster-rcnn/lib/nms/gpu_nms.cpp 重新编译即可
$ mv ~/py-faster-rcnn/lib/utils/bbox.c ~/py-faster-rcnn/lib/utils/bbox.c_rmbyfc
$ mv ~/py-faster-rcnn/lib/nms/cpu_nms.c ~/py-faster-rcnn/lib/nms/cpu_nms.c_rmbyfc
$ mv ~/py-faster-rcnn/lib/nms/gpu_nms.cpp ~/py-faster-rcnn/lib/nms/gpu_nms.cpp_rmbyfc
重新编译~/py-faster-rcnn/lib/里的文件
再
#./demo.py --net zf
终于成功了。
运行成功后,会有如下显示
