Ubuntu上编译Caffe和拓展应用(faster-rcnn, pvanet)的错误及解决方案

Caffe

错误： 采用make方式编译时遇到如下错误

In file included from /usr/include/boost/python/detail/prefix.hpp:13:0,
                 from /usr/include/boost/python/args.hpp:8,
                 from /usr/include/boost/python.hpp:11,
                 from tools/caffe.cpp:2:
/usr/include/boost/python/detail/wrap_python.hpp:50:23: fatal error: pyconfig.h: No such file or directory
compilation terminated.
Makefile:575: recipe for target '.build_release/tools/caffe.o' failed
make: *** [.build_release/tools/caffe.o] Error 1

解决方案：修改Makefile.config，将

PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
#                 $(ANACONDA_HOME)/include/python2.7 \
#                 $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include \

取消以下2行注释

PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
                 $(ANACONDA_HOME)/include/python2.7 \
                 $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include \

Note:$(ANACONDA_HOME) #虚拟环境Anaconda2的根目录

Faster-RCNN

问题： 如何编译只采用cpu版本的Faster-RCNN?

解决方案：
在./lib/setup.py中注释以下部分

...
#CUDA = locate_cuda()
...
...
#self.set_executable('compiler_so', CUDA['nvcc'])
...
...
#Extension('nms.gpu_nms',
#[‘nms/nms_kernel.cu', 'nms/gpu_nms.pyx'],
#library_dirs=[CUDA['lib64']],
#libraries=['cudart'],
#language='c++',
#runtime_library_dirs=[CUDA['lib64']],
## this syntax is specific to this build system
## we're only going to use certain compiler args with nvcc and not with
## gcc the implementation of this trick is in customize_compiler() below
#extra_compile_args={'gcc': ["-Wno-unused-function"],
#’nvcc': ['-arch=sm_35',
#’—ptxas-options=-v',
#’-c’,
#’—compiler-options',
#”’-fPIC'"]},
#include_dirs = [numpy_include, CUDA['include']]
#)

问题：运行时，遇到错误：ImportError: No module named cv2

File "./tools/test_net.py", line 13, in <module>
    from fast_rcnn.test import test_net
  File "/home/rtc5/JpHu/pva-faster-rcnn-master/tools/../lib/fast_rcnn/test.py", line 15, in <module>
    import cv2
ImportError: No module named cv2

解决方案
（1）检查cv2是否存在:
在${HOME}目录下运行

$find -name cv2

进行查找
（2）如果不存在cv2，安装python-opencv

sudo apt-get install python-opencv

（3）如果存在cv2,将文件夹cv2所在目录添加到.bashrc最后一行（如我将cv2安装在/home/rtc5/anaconda2/envs/tensorflow/lib/python2.7/site-packages/cv2下）

export PATHONPATH=$PYTHONPATH:/home/rtc5/anaconda2/envs/tensorflow/lib/python2.7/site-packages/cv2

运行命令

source ./bashrc #激活

激活./bashrc

问题：编译cpu版本成功后，faster-rcnn运行时，遇到错误ImportError: No module named gpu_nms

File "./demo.py", line 18, in
from fast_rcnn.test import im_detect
File ".../py-faster-rcnn-master/tools/../lib/fast_rcnn/test.py", line 17, in
from fast_rcnn.nms_wrapper import nms
File ".../py-faster-rcnn-master/tools/../lib/fast_rcnn/nms_wrapper.py", line 11, in
from nms.gpu_nms import gpu_nms
ImportError: No module named gpu_nms

解决方案：
注释${FCNN}/py-faster-rcnn/lib/fast_rcnn/nms_wrapper.py 中有关gpu的代码

from fast_rcnn.config import cfg
#from nms.gpu_nms import gpu_nms
from nms.cpu_nms import cpu_nms

def nms(dets, thresh, force_cpu=False):
    """Dispatch to either CPU or GPU NMS implementations."""

    if dets.shape[0] == 0:
        return []
    #if cfg.USE_GPU_NMS and not force_cpu:
    #    return gpu_nms(dets, thresh, device_id=cfg.GPU_ID)
    else:
        return cpu_nms(dets, thresh)

问题：（1）运行vgg16版本的faster-rcnn的./tools/demo.py遇到如下问题

WARNING: Logging before InitGoogleLogging() is written to STDERR
F1207 00:08:31.251930 20944 common.cpp:66] Cannot use GPU in CPU-only Caffe: check mode.
*** Check failure stack trace: ***
Aborted (core dumped)

解决方案：
采用命令：

$./tools/demo.py --cpu

Note:运行pvanet示例时，遇到类似问题，则需要将测试文件*.py中set_gpu的相关代码注释

问题：如何编译cpu版本的pvanet

编译caffe，遇到问题：

src/caffe/layers/proposal_layer.cpp:321:10: error: redefinition of ‘void caffe::ProposalLayer<Dtype>::Backward_gpu(const std::vector<caffe::Blob<Dtype>*>&, const std::vector<bool>&, const std::vector<caffe::Blob<Dtype>*>&)’
 STUB_GPU(ProposalLayer);
          ^
./include/caffe/util/device_alternate.hpp:17:6: note: in definition of macro ‘STUB_GPU’
 void classname<Dtype>::Backward_gpu(const vector<Blob<Dtype>*>& top, \
      ^
In file included from src/caffe/layers/proposal_layer.cpp:1:0:
./include/caffe/fast_rcnn_layers.hpp:122:16: note: ‘virtual void caffe::ProposalLayer<Dtype>::Backward_gpu(const std::vector<caffe::Blob<Dtype>*>&, const std::vector<bool>&, const std::vector<caffe::Blob<Dtype>*>&)’ previously declared here
   virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
                ^
Makefile:575: recipe for target '.build_release/src/caffe/layers/proposal_layer.o' failed
make: *** [.build_release/src/caffe/layers/proposal_layer.o] Error 1
make: *** Waiting for unfinished jobs....

解决方案：
由于caffe::ProposalLayer<Dtype>::Backward_gpu在./include/caffe/fast_rcnn_layers.hpp和./include/caffe/util/device_alternate.hpp（后者为模板形式）中定义了两次，被系统认为重定义。
解决方法如下：
将./include/caffe/fast_rcnn_layers.hpp的Backward_gpu代码

 virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom){}

修改如下

 virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
      const vector<bool>& propagate_down, const vector<Blob<Dtype>*>& bottom);

由于Backward_cpu只在./include/caffe/fast_rcnn_layers.hpp中定义过一次，所以一定避免对它做如上gpu的修改。

问题：如何只用cpu训练caffe,py-faster-rcnn,pvanet?

*报错：

smooth_L1_loss_layer Not Implemented Yet

解决方案：*
补充./src/caffe/layers/smooth_L1_loss_layer.cpp函数实体SmoothL1LossLayer::Forward_cpu和SmoothL1LossLayer::Backward_cpu

`// ------------------------------------------------------------------
// Fast R-CNN
// Copyright (c) 2015 Microsoft
// Licensed under The MIT License [see fast-rcnn/LICENSE for details]
// Written by Ross Girshick
// ------------------------------------------------------------------

#include "caffe/fast_rcnn_layers.hpp"

namespace caffe {

template
void SmoothL1LossLayer::LayerSetUp(
const vector<Blob>& bottom, const vector<Blob>& top) {
SmoothL1LossParameter loss_param = this->layer_param_.smooth_l1_loss_param();
sigma2_ = loss_param.sigma() * loss_param.sigma();
has_weights_ = (bottom.size() >= 3);
if (has_weights_) {
CHECK_EQ(bottom.size(), 4) << "If weights are used, must specify both "
"inside and outside weights";
}
}

template
void SmoothL1LossLayer::Reshape(
const vector<Blob>& bottom, const vector<Blob>& top) {
LossLayer::Reshape(bottom, top);
CHECK_EQ(bottom[0]->channels(), bottom[1]->channels());
CHECK_EQ(bottom[0]->height(), bottom[1]->height());
CHECK_EQ(bottom[0]->width(), bottom[1]->width());
if (has_weights_) {
CHECK_EQ(bottom[0]->channels(), bottom[2]->channels());
CHECK_EQ(bottom[0]->height(), bottom[2]->height());
CHECK_EQ(bottom[0]->width(), bottom[2]->width());
CHECK_EQ(bottom[0]->channels(), bottom[3]->channels());
CHECK_EQ(bottom[0]->height(), bottom[3]->height());
CHECK_EQ(bottom[0]->width(), bottom[3]->width());
}
diff_.Reshape(bottom[0]->num(), bottom[0]->channels(),
bottom[0]->height(), bottom[0]->width());
errors_.Reshape(bottom[0]->num(), bottom[0]->channels(),
bottom[0]->height(), bottom[0]->width());
// vector of ones used to sum
ones_.Reshape(bottom[0]->num(), bottom[0]->channels(),
bottom[0]->height(), bottom[0]->width());
for (int i = 0; i < bottom[0]->count(); ++i) {
ones_.mutable_cpu_data()[i] = Dtype(1);
}
}

template
void SmoothL1LossLayer::Forward_cpu(const vector<Blob>& bottom,
const vector<Blob>& top) {
// NOT_IMPLEMENTED;
int count = bottom[0]->count();
//int num = bottom[0]->num();
const Dtype* in = diff_.cpu_data();
Dtype* out = errors_.mutable_cpu_data();
caffe_set(errors_.count(), Dtype(0), out);

caffe_sub(
count,
bottom[0]->cpu_data(),
bottom[1]->cpu_data(),
diff_.mutable_cpu_data()); // d := b0 - b1
if (has_weights_) {
// apply "inside" weights
caffe_mul(
count,
bottom[2]->cpu_data(),
diff_.cpu_data(),
diff_.mutable_cpu_data()); // d := w_in * (b0 - b1)
}

for (int index = 0;index < count; ++index){
Dtype val = in[index];
Dtype abs_val = abs(val);
if (abs_val < 1.0 / sigma2_) {
out[index] = 0.5 * val * val * sigma2_;
} else {
out[index] = abs_val - 0.5 / sigma2_;
}
}

if (has_weights_) {
// apply "outside" weights
caffe_mul(
count,
bottom[3]->cpu_data(),
errors_.cpu_data(),
errors_.mutable_cpu_data()); // d := w_out * SmoothL1(w_in * (b0 - b1))
}

Dtype loss = caffe_cpu_dot(count, ones_.cpu_data(), errors_.cpu_data());
top[0]->mutable_cpu_data()[0] = loss / bottom[0]->num();
}

template
void SmoothL1LossLayer::Backward_cpu(const vector<Blob>& top,
const vector& propagate_down, const vector<Blob>& bottom) {
// NOT_IMPLEMENTED;
int count = diff_.count();
//int num = diff_.num();
const Dtype* in = diff_.cpu_data();
Dtype* out = errors_.mutable_cpu_data();
caffe_set(errors_.count(), Dtype(0), out);

for (int index = 0;index < count; ++index){
Dtype val = in[index];
Dtype abs_val = abs(val);
if (abs_val < 1.0 / sigma2_) {
out[index] = sigma2_ * val;
} else {
out[index] = (Dtype(0) < val) - (val < Dtype(0));
}
}

for (int i = 0; i < 2; ++i) {
if (propagate_down[i]) {
const Dtype sign = (i == 0) ? 1 : -1;
const Dtype alpha = sign * top[0]->cpu_diff()[0] / bottom[i]->num();
caffe_cpu_axpby(
count, // count
alpha, // alpha
diff_.cpu_data(), // x
Dtype(0), // beta
bottom[i]->mutable_cpu_diff()); // y
if (has_weights_) {
// Scale by "inside" weight
caffe_mul(
count,
bottom[2]->cpu_data(),
bottom[i]->cpu_diff(),
bottom[i]->mutable_cpu_diff());
// Scale by "outside" weight
caffe_mul(
count,
bottom[3]->cpu_data(),
bottom[i]->cpu_diff(),
bottom[i]->mutable_cpu_diff());
}
}
}
}

#ifdef CPU_ONLY
STUB_GPU(SmoothL1LossLayer);
#endif

INSTANTIATE_CLASS(SmoothL1LossLayer);
REGISTER_LAYER_CLASS(SmoothL1Loss);

} // namespace caffe

转自： zhouphd 的解答，已验证有效，caffe能够通过编译，并进行训练

问题：运行pvanet时，报错

Intel MKL FATAL ERROR: Cannot load libmkl_avx2.so or libmkl_def.so.

原因：由于之前安装tensorflow时，采用的是anaconda，它独自创建了一个虚拟环境（自行另安装依赖库），但由于anaconda会在~/.bashrc中添加PATH路径。所以执行caffe程序时（在虚拟环境之外），其依赖库也会受到anaconda安装软件的影响。
解决方案：屏蔽anaconda设置的PATH，在~/.bashrc中注释

#export PATH="/home/cvrsg/anaconda2/bin:$PATH"
$source ~/.bashrc #激活.bashrc

注意：重开一个终端，在当前终端，source命令是没有生效的。
如何验证？

如果在当前终端下输入
sudo echo $PATH
你会发现anaconda2/bin还在PATH中，source未生效
重开终端之后，
anaconda2/bin已消失

同样由此可知，当我们需要anaconda2时，我们可以将

#export PATH="/home/cvrsg/anaconda2/bin:$PATH"

解注释，并source ~/.bashrc激活
不需要时，注释即可。
在上述命令被注释的情况下，运行source activate tensorflow会出现以下错误

bash: activate: No such file or directory

别着急，解注释，激活就好。

Note-切记：：：
另外，如果我们要用conda安装软件时，一定要切换到相应的虚拟环境下，否则安装的软件很容易和系统软件发生版本冲突，导致程序出错。

在安装pycaffe依赖库时，遇到的问题

利用命令for req in $(cat requirements.txt); do pip install $req; done安装pycaffe相关依赖库遇到问题：ImportError: No module named packaging.version
描述：这是因为采用 sudo apt-get install python-pip安装的pip有问题

sudo apt-get remove python-pip #删除原有pip
wget https://bootstrap.pypa.io/get-pip.py  //获取特定pip，并进行安装
sudo python get-pip.py

错误

F0608 15:36:07.750129  6353 concat_layer.cpp:42] Check failed: top_shape[j] == bottom[i]->shape(j) (63 vs. 62) All inputs must have the same shape, except at concat_axis.
*** Check failure stack trace: ***
Aborted (core dumped)

这个似乎是新版本的PVANET的问题，旧版本的PVANET没有该问题。问题出在lib文件的改变。

其他

问题: wget如何避免防火墙的影响？

解决方案：
在命令

wget xxx
#如wget https://www.dropbox.com/s/87zu4y6cvgeu8vs/test.model?dl=0 -O models/pvanet/full/test.model

之后加

—no-check-certificate